You are on page 1of 184

SHRP-A-407

The Superpave Mix Design Manual


for New Construction and Overlays

Ronald J. Cominsky
University of Texas at Austin

With contributions by:

Gerald A. Huber
Heritage Research Group

Thomas W. Kennedy
University of Texas at Austin

Michael Anderson
The Asphalt Institute

Strategic Highway Research Program


National Research Council
Washington, DC 1994
SHRP-A-407
Contract A-001
ISBN 0-309-05804-X
Product No.: 1012

Program Manager: Edward T. Harrigan


Program Area Secretary: Juliet Narsiah
Production Editor: Michael Jahr

May 1994

key words:
aggregate
asphalt binder
asphalt cement
asphalt modifier
bending beam rheometer
direct tension
dynamic shear rheometer
environmental conditioning system
gyratory compaction
hot mix asphalt
indirect tensile test device
net adsorption
performance grade
pressurized aging vessel
rolling wheel compaction
shear test device
SUPERPAVE

Strategic Highway Research Program


National Research Council
2101 Constitution Avenue N.W.
Washington, DC 20418

(202) 334-3774

The publication of this report does not necessarily indicate approval or endorsement of the findings, opinions,
conclusions, or recommendations either inferred or specifically expressed herein by the National Academy of
Sciences, the United States Government, or the American Association of State Highway and Transportation
Officials or its member states.

1994 National Academy of Sciences

1.5M/NAP/594
Acknowledgments

The research described herein was supported by the Strategic Highway Research Program
(SHRP). SHRP is a unit of the National Research Council that was authorized by section
128 of the Surface Transportation and Uniform Relocation Assistance Act of 1987.

.
in
Table of Contents

Acknowledgments ................................................. iii

List of Figures ..................................................... xi

List of Tables .................................................... xv

1. Introduction

1.1 The Purpose of This Manual ................................. 1


1.2 The Superpave Mix Design System ............................ 1

1.2.1 Concept of the Superpave Mix Design ..................... 2


1.2.2 Low-, Intermediate-, and High-Traffic Level Designs ........... 4

1.2.2.1 Level 1 (Low Traffic) Mix Design .................. 5


1.2.2.2 Level 2 (Intermediate Traffic) Mix Design ............ 5
1.2.2.3 Level 3 (High Traffic) Mix Design ................. 8

1.3 The Superpave Performance-Based Test Methods for Paving Mixes .... 10

1.3.1 Laboratory Compaction ............................... 11


1.3.2 Test Procedures for Moisture Sensitivity ................... 11
1.3.3 Conditioning Procedure for Aging ....................... 12
1.3.4 Performance Testing with the Shear Test Device ............. 12

1.3.4.1 Nonlinear Elastic Behavior ...................... 13


1.3.4.2 Viscoelastic Behavior .......................... 13
1.3.4.3 Tertiary Creep ............................... 13
1.3.4.4 Rutting Potential (Optional) ..................... 14
1.3.4.5 Performance Testing with the Indirect
Tensile Test Device ........................... 14

1.4 Interaction with the Superpave Software ........................ 15


1.5 Use of Modified Materials ................................. 17
1.6 Multi-Layer Designs in the Superpave System ................... 18

V
1.7 Ground Rules for Use of the Superpave Mix Design System ......... 19

1.7.1 Accounting for the Effects of Moisture Sensitivity ............ 19


1.7.2 Accounting for the Effects of Aging ...................... 20
1.7.3 Superpave Mix Designs for New Construction ............... 20
1.7.4 Superpave Mix Designs for HMA Overlays ................. 21
1.7.5 Accounting for Reflection Cracking ...................... 22
1.7.6 Units of Measurement in the Superpave System .............. 23
1.7.7 Pavement Structural Factors in Superpave ................. 23

2. Selecting Mixture Materials ..................................... 25

2.1 Aggregate ............................................. 25

2.1.1 Gradation Control ................................... 27


2.1.2 Coarse Aggregate Angularity ........................... 32
2.1.3 Fine Aggregate Angularity ............................. 33
2.1.4 Toughness ........................................ 33
2.1.5 Soundness ........................................ 34
2.1.6 Deleterious Materials ................................ 34
2.1.7 Clay Content ...................................... 34
2.1.8 Thin, Elongated Particles .............................. 35
2.1.9 Dust Proportion .................................... 35

2.2 Asphalt Binders

2.2.1 General Treatment of the Asphalt Binder


in the Superpave System ............................. 36
2.2.2 Selecting an Asphalt Binder
Performance Grade Based upon High and
Low Pavement Design Temperatures and Traffic Level ........ 37
2.2.3 Asphalt-Aggregate Compatibility with the
Net Adsorption Test (Optional) ......................... 48

2.3 Modifiers ............................................. 50


2.4 Next Steps in the Design Process ............................ 52

3 Level 1 Mix Design (Low Traffic Levels -- Volumetric Design) ............ 53

3.1 Introduction ........................................... 53


3.2 Volumetric Properties ..................................... 53

3.2.1 Definitions ....................................... 55


3.2.2 Air Voids ........................................ 55

vi
3.2.3 Voids in the Mineral Aggregate (VMA) ................... 57
3.2.4 Voids Filled with Asphalt (VFA) ....................... 60

3.3 Outline of Level 1 (Volumetric) Design Method .................. 60


3.4 Equipment ............................................ 61
3.5 Gyratory Compaction ..................................... 63
3.6 Short-Term Aging of Selected Target Asphalt-Aggregate Blends ....... 67
3.7 Selecting the Design of Aggregate Structure (Skeleton) ............. 67

3.7.1 Aggregate Trial Blend Gradations ....................... 68


3.7.2 Calculate an Initial Trial Asphalt Content for each
Trial Aggregate Gradation ............................ 70
3.7.3 Compact Specimens of Each Trial Gradation ............... 75
3.7.4 Evaluate Compacted Trial Aggregate Gradations ............. 75

3.7.4.1 Shifting Trial Aggregate Gradation


Densification Curves .......................... 76
3.7.4.2 Determination of Estimated Design Parameters ........ 77
3.7.4.3 Possible Remedies When Trial Aggregate
Gradations Do Not Meet Superpave Criteria .......... 82

3.8 Selecting the Design Asphalt Content ......................... 83

3.8.1 Selecting of Asphalt Contents for Evaluation ............... 83


3.8.2 Compact Specimens at Four Asphalt Contents .............. 84
3.8.3 Select the Design Asphalt Content Corresponding
to an Air Voids Content of Four Percent at Ndesig n ............ 84
3.8.4 Compare Volumetric Properties to Design Criteria
at Ninit and Nmax .................................... 86

3.9 Evaluate Moisture Susceptibility of the Design Aggregate


Structure at the Design Asphalt Content ........................ 87

3.9.1 Evaluation of Moisture Susceptibility with AASHTO Method


of Test T 283 (Superpave Alternative 3) .................. 87
3.9.2 Evaluation of Moisture Susceptibility with SHRP Method
of Test M-006 (Superpave Alternative II) .................. 88

3.10 Adjusting Mixture to Meet Properties ......................... 88

3.10.1 Voids in Mineral Aggregate (VMA) ..................... 89


3.10.2 Voids Filled With Asphalt (VFA) ....................... 89
3.10.3 Retained Tensile Strength or Resilient Modulus ............. 89

vii
3.11 Final Outcome of a Superpave Level 1 Mix Design ............... 90

4 Level 2 Mix Design (Intermediate Traffic Levels) ...................... 91

4.1 Introduction ............................................. 91


4.2 Outline of Level 2 Mix Design Method ........................ 92

4.2.1 Historical Air Temperature ............................ 95


4.2.2 Historical Pavement Temperature ........................ 95
4.2.3 Pavement Temperature Converted Into Effective Temperature .... 95
4.2.4 Calculate Material Properties ........................... 95
4.2.5 Predict Pavement Performance .......................... 95

4.3 Equipment ............................................ 96


4.4 Selection of Mixture Materials for Volumetric Design .............. 97
4.5 Selection of Asphalt Contents to Yield
3, 4, and 6 Percent Air Voids ............................... 97
4.6 Determining the Effective Temperature for Fatigue Testing .......... 98
4.7 Determining the Effective Temperature for Permanent
Deformation Testing Including Tertiary Creep ................... 99
4.8 Tertiary Creep Evaluation ................................. 101
4.9 Permanent Deformation and Load-Associated
Fatigue Property Characterization ........................... 106
4.10 Predicting Permanent Deformation and
Load-Associated Fatigue Cracking at the
Pavement Design Life ................................... 107
4.11 Low-Temperature Cracking (Non-Load Associated)
Property Characterization ................................. 108
4.12 Predicting Low-Temperature Cracking at the
Pavement Design Life ................................... 108
4.13 Selecting the Optimum Asphalt Content ....................... 109
4.14 Treatment of Modified Materials and Mixtures .................. 111

5 Level 3 Mix Design (High Traffic Levels)

5.1 Introduction .......................................... 113


5.2 Outline of Level 3 Mix Design Method ....................... 115

5.2.1 Historical Air Temperature ........................... 115


5.2.2 Historical Pavement Temperature ....................... 116
5.2.3 Calculate Material Properties .......................... 116
5.2.4 Predict Pavement Performance ......................... 116


Vlll
5.3 Equipment ........................................... 116
5.4 Selecting Mixture Materials for
Volumetric Design ...................................... 119
5.5 Selecting Asphalt Contents to Yield
3, 4, and 6 Percent Air Voids .............................. 120
5.6 Determining Effective Temperature for
the Screening Test for Tertiary Creep ........................ 120
5.7 Tertiary Creep Mix Evaluation ............................. 120
5.8 Permanent Deformation and Load-Associated
Fatigue Cracking Property Characterization .................... 121
5.9 Predicting Permanent Deformation and
Load-Associated Fatigue Cracking at the
Pavement Design Life ................................... 122
5.10 Low-Temperature Cracking (Nonload Associated)
Property Characterization ................................. 123
5.11 Predicting Low-Temperature Cracking at the
Pavement Design Life ................................... 123
5.12 Selecting the Optimum Asphalt Content ....................... 124
5.13 Treatment Modified Materials and Mixtures .................... 126
5.14 Proof Testing ......................................... 128

5.14.1 Rolling Wheel Compaction .......................... 128


5.14.2 Wheel-Tracking Device ............................. 128
5.14.3 Flexural Beam Fatigue Test .......................... 128
5.14.4 Thermal Stress Restrained Specimen Test ................ 129

6 Superpave Mix Design: An Example .............................. 131

6.1 Select Materials ........................................ 131

6.1.1 Coarse Aggregate Angularity .......................... 134


6.1.2 Fine Aggregate Angularity ............................ 134
6.1.3 Thin and Elongated Particles .......................... 135
6.1.4 Clay Content (Sand Equivalent) ........................ 135

6.2 Select Design Aggregate Structure ........................... 136


6.3 Select Design Asphalt Binder Content ........................ 146
6.4 Evaluate Moisture Sensitivity .............................. 146
6.5 Superpave Level 3 Mixture Design Example .................... 147

6.5.1 Introduction ...................................... 147


6.5.2 Level 3 Mix Design and Pavement Performance
Prediction Approach ............................... 149

ix
6.5.2.1 Define Project Where Mixture is to be Used ......... 150
6.5.2.2 Select Asphalt Contents for Analysis .............. 150
6.5.2.3 Perform Repeated Shear Screening Test ............ 152

6.5.3 Performance-Based Tests and Results .................... 152


6.5.4 Material Properties and Material Property Model ............ 157
6.5.5 Environment Effects Model ........................... 157
6.5.6 Response and Distress Models to Predict Rutting ............ 159
6.5.7 Selection of the Optimum Asphalt Content ................ 160

Appendix A
Superpave Aggregate Gradation Control Points and Restricted Zones ............. 163

Appendix B
Superpave Specimen Numbers, Test Codes and Temperature Codes for
Level 2 and Level 3 Mix Designs ..................................... 167

Appendix C
Standard Method of Test for Determining the Percentage of Crushed Fragments
in Gravel ...................................................... 169

C. 1 Purpose and Scope ....................................... 169


C.2 Sampling .............................................. 169
C.3 Apparatus ............................................. 169
C.4 Procedure ............................................. 170
C.5 Calculations ............................................ 170
C.6 Report ................................................ 170
C.7 References ............................................. 170

Appendix D
Superpave Paving Mix Designation .................................... 171

X
List of Figures

Figure 1.1. Structure of the Superpave Mix Design System .................... 3

Figure 1.2. Level 1 Superpave Mix Design ............................... 6

Figure 1.3. Superpave Level 2 ........................................ 7

Figure 1.4. Superpave Level 3 ........................................ 9

Figure 1.5. Structure of the Superpave Specification,


Mix Design and Support Program ............................ 16

Figure 2.1. Federal Highway Administration 0.45 Aggregate Grading Chart ....... 28

Figure 2.2. Superpave Gradation Control Points and Restricted Zone


for a 12.5 mm Nominal Maximum Size Aggregate Gradation ......... 29

Figure 2.3a Performance Graded Asphalt Binder Specification (AASHTO MPI) .... 38

Figure 2.3b Performance Graded Asphalt Binder Specification (Continued) ........ 39

Figure 2.4. Typical Probability Distribution of the 7-Day Average


Maximum Pavement Temperature ............................ 41

Figures 2.5a and b Example 1: Probabilities for a Mean of 57C and a


Standard Deviation of 2C ............................ 42

Figures 2.6a and b Example 2: Probabilities for a Mean of 59C and a


Standard Deviation of 2C ............................ 44

Figure 2.7. Example 3: Probabilities for a Mean of 61C and a Standard


Deviation of 2C ........................................ 45

Figure 2.8. Selection of Asphalt Binder Performance Grades on the Basis


of Climate, Traffic Speed, and Traffic Volume ................... 47

xi
Figure 3.1. Level 1 Superpave Mix Design .............................. 54

Figure 3.2. Definitions of the Volumetric Properties of Compacted Asphalt


Paving Mix ............................................ 56

Figure 3.3. Minimum Percent Voids in Mineral Aggregate (VMA) for 3, 4, and 5
Percent Air Voids ....................................... 59

Figure 3.4. Typical Densification Curve Obtained With SHRP Gyratory Compactor 64

Figure 3.5. Gyratory Compaction Densification Curves for Four Asphalt


Binder Contents ......................................... 69

Figure 3.6. Evaluation of the Aggregate Gradations of Three Trial Blends ........ 71

Figure 3.7. Calculating the Slope of a Densification Curve ................... 78

Figure 3.8. Shifting the Densification Curve of a Trial Asphalt Content .......... 78

Figure 3.9. Plots of Percent Air Voids, Percent VMA, Percent VFA
and Density Versus Percent Asphalt Content ..................... 85

Figure 4.1. Superpave Level 2 Mix Design .............................. 93

Figure 4.2. Typical Relationship of Plastic (Permanent) Strain, ep, and Load
Repetitions, N, for Mixes with Tertiary Creep (Plastic Flow) ........ 102

Figure 4.3. Selection of Approximate Control Temperature (To) Curve .......... 103

Figure 4.4. Selection of the Approximate Equivalent Laboratory Repetition Curve 105

Figure 4.5. Selection of Optimum Asphalt Content Range Satisfying


Three Distress Factor Conditions ............................ 110

Figure 4.6. Selection of Optimum Asphalt Content Range Satisfying


Two Distress Factor Conditions ............................. 112

Figure 5.1. Superpave Level 3 Mix Design ............................. 114

Figure 5.2. Selection of an Optimum Asphalt Content Range Satisfying


Three Distress Factor Conditions ............................ 125

Figure 5.3. Selection of an Optimum Asphalt Content Range Satisfying


Two Distress Factor Conditions ............................. 127

xii
Figure 6.1. Superpave Gradation Control Points and Restricted Zone
for a 19.0 mm Nominal Maximum Size Aggregate Gradation ........ 138

Figure 6.2. Aggregate Gradations of Three Trial Blends .................... 140

Figure 6.3. Flow Diagram of the Different Parts of the Superpave Pavement
Performance Model ..................................... 148

Figure 6.4. FHWA Environmental Effects Model Map of U.S. Climatic Regions . . . 151

Figure 6.5. Selection of an Optimum Asphalt Content Range Satisfying


Permanent Deformation (Rutting) Requirements ................. 161

,
Xlll
List of Tables

Table 1.1. Recommended Design Traffic for Level 1, 2, and 3 Mix Designs ....... 4

Table 1.2. Superpave Mix Design Tests and Equipment .................... 10

Table 2.1. Suggested Maximum Aggregate Sizes ......................... 26

Table 2.2. Definition of Aggregate Fractions ............................ 27

Table 2.3. Definition of Aggregate Types .............................. 27

Table 2.4. Aggregate Gradation Control Points ........................... 30

Table 2.5. Common Sieves Sizes Used by State Highway Agencies ............ 31

Table 2.6. Coarse Aggregate Angularity Criteria .......................... 32

Table 2.7. Fine Aggregate Angularity Criteria ........................... 33

Table 2.8. Clay Content Criteria ..................................... 35

Table 2.9. Criteria For Thin, Elongated Particles ......................... 35

Table 2.10. Criteria For Dust Proportion ................................ 36

Table 2.11. Suggested Net Adsorption Criteria ............................ 49

Table 3.1. Air Voids Criteria ....................................... 57

Table 3.2. Voids in Mineral Aggregate Criteria .......................... 58

Table 3.3. Voids Filled With Asphalt Criteria ........................... 60

Table 3.4. Number of Initial (N_), Design (Nd) and Maximum (Nm) Gyrations
Required For Various Traffic Levels and Maximum Temperature
Environments .......................................... 63

XV
Table 3.5. General Superpave Compaction Requirements .................... 65

Table 3.6. Calculation of AVMA .................................... 80

Table 3.7. Summary of Volumetric Design Criteria ........................ 80

Table 3.8. Selection of a Design Aggregate Structure ...................... 81

Table 3.9. Sample Volumetric Design Data at Ndesign ....................... 86

Table 4.1. Performance Tests For Level 2 Mix Design ..................... 94

Table 4.2. Compacted Specimen Requirement for Level 2 Mix Design .......... 94

Table 4.3. Values of K_ .......................................... 100

Table 4.4. Maximum Shear and Compressive Stress Levels ................. 104

Table 4.5. Suggested Acceptable Levels of Performance Level 2


Mix Design ........................................... 109

Table 5.1. Performance Tests for Level 3 Mix Design ..................... 117

Table 5.2. Numbers of Compacted Specimen Required for Level 3


Mix Design (Per Asphalt Content) ........................... 118

Table 5.3. Suggested Acceptable Levels of Performance Requirements


(Level 3 Mix Designs) ................................... 124

Table 6.1. Project Weather Data .................................... 132

Table 6.2. Binder Specification Test Results ........................... 133

Table 6.3. Aggregate Specific Gravities ............................... 133

Table 6.4. Coarse Aggregate Angularity .............................. 134

Table 6.5. Fine Aggregate Angularity ................................ 134

Table 6.6. Thin and Elongated Particles ............................... 135

Table 6.7. Clay Content (Sand Equivalent) ............................ 135

Table 6.8. Gradation Criteria for 19.0 mm Nominal Maximum Size Aggregate . . . 136

xvi
Table 6.9. Estimated Aggregate Blend Properties ........................ 137

Table 6.10. Aggregate Blending Table for Trial Blends 1, 2 and 3 ............. 139

Table 6.11. Summary of Trial Blends ................................. 144

Table 6.12. Estimated Values for Trial Blends ........................... 144

Table 6.13. Mixture Criteria ........................................ 145

Table 6.14. Dust-to-Asphalt Ratio of Trial Blends ........................ 145

Table 6.15. Comparison of Trial Blends ............................... 145

Table 6.16. Design Mix Properties @ 4.7% Asphalt Content ................. 146

Table 6.17. Moisture Susceptibility of Design Asphalt Blend


by AASHTO T 283 ..................................... 147

Table 6.18. Example Output Data from the Frequency Sweep Test ............. 154

Table 6.19. Example Test Data From Simple Shear at Constant Height Test ...... 155

Table 6.20. Example Test Data from the Volumetric Test ................... 156

Table 6.21. Example Data from the Uniaxial Strain Test .................... 156

Table 6.22. Examples of Material Property Parameters Calculated from


Superpave Material Property Model .......................... 158

Table 6.23. Average Pavement Temperatures at Depth Calculated by the


Environmental Effects Model .............................. 159

Table 6.24. Example of Predicted Rut Depth Data from Distress Model ......... 160

Table A-1. 37.5 mm Nominal Maximum Size ........................... 163

Table A-2. 25.0 mm Nominal Maximum Size ........................... 163

Table A-3. 19.0 mm Nominal Maximum Size ........................... 163

Table A-4. 12.5 mm Nominal Maximum Size ........................... 164

Table A-5. 9.5 mm Nominal Maximum Size ............................ 164

xvii
'Fable A-6. Boundaries of Aggregate Restricted Zone ...................... 165

"Fable B-1. Requirements for Superpave Level 2 Mix Design ................ 167

Table B-2. Requirements for Superpave Level 3 Mix Design ................ 168

Table D- 1. Traffic Level Designator .................................. 171

Table D-2. Gradation Designator .................................... 171

Table D-3. Nominal Maximum Size Designator .......................... 172

Table D-4. Paving Mix Depth Designator .............................. 172

xviii
1 Introduction

1.1 The Purpose of This Manual

This laboratory manual presents the Superpave mix design system in a complete,
step-by-step format. It is intended for engineers and technicians in public and private
organizations to use when designing paving mixes for all classes of highways, from farm-
to-market roads to urban freeways.

An essential companion to this manual is The Superpave Mix Design System Manual
of Specifications, Test Methods and Practices.

The Superpave software program (The Superpave Specification, Mix Design and
Support Program) and its users manual are also necessary to take full advantage of the mix
design system. This software is designed to run on an 80386-based or, preferably, an
80486-based personal computer.

1.2 The Superpave Mix Design System

The Superpave mix design system is a comprehensive method of designing paving


mixes tailored to the unique performance requirements dictated by the traffic, environment
(climate), and structural section at a particular pavement site. It facilitates selecting and
combining asphalt binder, aggregate, and any necessary modifier to achieve the required
level of pavement performance.

The Superpave system is applicable to virgin and recycled, dense-graded, hot mix
asphalt (HMA), with or without modification. In addition, the Superpave performance tests
are applicable to the characterization of a variety of specialized paving mixes such as stone
matrix asphalt (SMA). It can be used when constructing new surface, binder, and base
layers, as well as overlays on existing pavements. Through materials selection and mix
design, it directly addresses the reduction and control of permanent deformation, fatigue
cracking, and low-temperature cracking. It also explicitly considers the effects of aging and
moisture sensitivity in promoting or arresting the development of these three distresses.

The objective of the Superpave mix design system is to define an economical blend
of asphalt binder and aggregate that yields a paving mix having
* sufficient asphalt binder,

* sufficient voids in the mineral aggregate (VMA) and air voids,

sufficient workability, and

,, satisfactory performance characteristics over the service life of the pavement.

The Superpave mix design system has several distinctive features. First, only
performance-based and performance-related properties are used as the criteria for selection
of a mix design. Performance-based properties directly govern the response of the
pavement to load; performance can be predicted from this response. Performance-related
properties are those which are indirectly related to performance. They affect performance,
but do not, in themselves, control it.

As the traffic and environmental demands on the pavement increase, the Superpave
mix design system relies more on performance-based properties to select the optimum mix
design. The maximum possible contribution of the materials in the paving mix to pavement
performance can be achieved regardless of its structure.

Second, the Superpave performance-based specification for asphalt-aggregate


mixtures is based on the predicted performance of the pavement built with the paving mix
under design. The Superpave system provides a uniquely tailored specification, expressed
in terms of the rut depth, area of fatigue cracking, and spacing of low-temperature cracking
expected over a selected service life, for any paving project with significant traffic and
environmental demands. These specifications may be compared to criteria presented in this
manual or to criteria established by agency policy for discrete classes of traffic.

Third, in determining pavement performance, Superpave explicitly considers the


interaction of pavement structure, traffic, and environment with the paving mix. Thus, the
mix design and structural design can be integrated into an single system.

Finally, the design method provides a truly objective measure of the benefits or
penalties associated with the use of materials of varying levels of quality.

1.2.1 Concept of the Superpave Mix Design

The concept behind the Superpave mix design system is straightforward: use
available materials to prepare a mix design that achieves a level of performance
commensurate with the demands of traffic, environment, structure, and reliability (or,
conversely, risk) on the pavement.

This concept is illustrated by the general flow of a Superpave mix design (figure
1.1). At each step in the design, from materials selection through volumetric design and

2
Figure 1.1. Structure of the Superpave Mix Design System

Z
u') I AGGREGATESELECTION 1 I ASPHALT SELECTION )
LIJ

k-

DESIGN INCLUDE I
"J
O
ILl_ ( MOISTURE SUSCEPTIBILITY.,)
VOLUMETRIC MIXTURE

PERFORMANCEBASED
MATERIAL PROPERTIES
MEASUREMENT OF 1

(/)
I,i
t-
rr r I _
W
n ESTIMATE OF PAVEMENT ESTIMATE OF PAVEMENT
O PERFORMANCE PERFORMANCE
rr
n
_J
O
Z
"1-
O
LIJ

--I
I

l -- 4 i i

O FINAL MIX DESIGN OPTIONAL


n."
I- FOR PRODUCTION
Z

J
w
IT"
I FIELD MIXTURE
CONTROL TESTS
1
'_"LEVEL 3 PROVIDES
THE HIGHESTRELIABLEESTIMATEOF PAVEMENTPERFORMANCE 3
engineering evaluation of trial mix designs to field control, decisions must be made on the
basis of satisfying the specific pavement performance requirements to the extent possible. _

1.2.2 Low-, Intermediate-, and High-Traffic Level Designs

The Superpave mix design system contains three distinct levels of design, termed
level 1, level 2, and level 3. This feature permits the agency to select a design process that
is appropriate for the traffic loads and volume (expressed as total 80 kN equivalent single
axle loads (ESALs) over the service life of the pavement) expected for the paving project.
General recommendations for applying the three design levels are presented in table 1.1.

Table 1.1. Recommended Design Traffic For Level 1, 2, and 3 Mix Designs

Design Level Design Traffic (80 kN ESALs)


1 (low) _<106
2 (intermediate) _<107
3 (high) > 107

These traffic levels are presented as suggested guidelines only. Higher or lower
transition points may be set on the basis of individual agency policy.

In addition, all three design levels explicitly consider the effects of the climate
(environment) on pavement performance. Selection of the performance grade of asphalt
binder is guided by the high and low pavement design temperatures at the paving project,
the traffic speeds, and the design traffic levels. Candidate paving mixes are evaluated for
unacceptable moisture sensitivity. Both asphalt binders and candidate paving mixes are aged
in the laboratory to simulate the effects of both short- and long-term aging in pavement
service.

The complexity of the design process increases significantly from level 1 to level 3.
Level 3 requires a greater number of tests, more test specimens, and more time to complete
a design. In return, the reliability of the design -- that is, the probability that the paving

_By contrast, the widely used Marshall and Hveem methods of mix design are neither
performance-based nor performance-related. They are concerned primarily with achieving a
mix design with a stable, economical balance of aggregate and asphalt binder that features
sufficient workability to permit efficient placement of the mix. Both methods attempt to
gauge anticipated performance with empirical properties, such as Marshall stability and
flow, but neither method can ensure that a trial mix design will meet specific pavement
performance criteria.

4
mix will provide satisfactory pavement service under the anticipated conditions of traffic
and climate-increases proportionally.

The Superpave mix design system provides flexibility to deal with one, two, or all
three of the major distress types which it addresses. As examples, at the discretion of the
agency, designs for warm weather climates can concentrate on permanent deformation only
or on permanent deformation and fatigue cracking. In extreme cold weather climates,
designs can be aimed exclusively toward preventing low-temperature cracking without
regard for the development of other distresses. In a climate that experiences extremes of
both heat and cold, the agency can choose to use a level 3 design for permanent
deformation, while addressing fatigue and low-temperature cracking with the level 2 design
process.

1.2.2.1. Level 1 (Low Traffic) Mix Design

The Superpave level 1 mix design is presented as a flow chart in figure 1.2 and is
described in detail in chapters 2 and 3 of this manual.

Level 1 mix design employs a performance-based asphalt binder specification with


empirical, performance-related aggregate specifications, and principles of volumetric mix
design to obtain a paving mix with satisfactory performance for low-traffic paving projects
without the need for performance-based testing. Final selection of the design asphalt content
is based upon attaining specified levels of air voids, voids in mineral aggregate, and voids
filled with asphalt at initial, design, and maximum levels of compaction.

It is not possible to estimate the pavement performance of level 1 mix designs with
regard to permanent deformation, fatigue cracking, or low-temperature cracking without the
level 2 or 3 performance-based tests. However, the level 1 mix design provides a
reasonable guarantee of adequate performance if all of the specified criteria are met.

Gyratory compaction is the key to successful level 1 mix designs. In addition, this
design level explicitly considers the effects of moisture sensitivity and aging in selecting the
final mix design.

1.2.2.2 Level 2 (Intermediate Traffic) Mix Design

The Superpave level 2 mix design is presented as a flow chart in figure 1.3 and is
described in detail in chapter 4 of this manual.

The level 2 mix design incorporates the selection of a design asphalt content with
the volumetric (level 1) design procedure. Candidate mixes prepared at the design asphalt
content and at a high and a low asphalt content bracketing the design value are subjected to
a series of performance-based tests selected for use in routine mix designs.
Figure 1.2. Level 1 Superpave Mix Design (Note: All specimens are compacted from
paving mix that has been short-term aged (SHRPM-007)).

Select aggregate based upon Select asphalt binder grades


specification criteria based upon high and low
- gradation limits pavement design temperatures. (1-2)
- coarse aggregate
angul ari ty I
- fine aggregate
angularity Conduct optional net adsorption
clay content test to evaluate asphalt-aggregate
thin and elongated compatibility. (2 specimens) (1-2A)
particles (1-1) .......................................

IPrepare three or four trial


I gradations
determined from control points
established on 0.45 power curve. (I-3)

Calculate an initial trial asphalt


content for each trial gradation
(Two specimens compacted to N_esig n
per gradation = 6 or 8 specimens) (1-4)

Evaluate compacted trial gradations; select


a design aggregate structure; calculate an
estimated design asphalt content. (1-5)

Determine moisture Compact specimens at 4 Aes around


susceptibility of design the estimated design asphalt content
asphalt content and design (2 compacted specimens per AC = 8
gradation using AASHTOT 283 specimens) (I-6)
(Compact 6 specimens at 7%
AV. Test three dry and
three wet.) (1-8)
I
I
Select the design AC that satisfies
the volumetric criteri a at N design
and the Nine
x and Ni_it
criteria (1-7)

,e,
I 4__,_'::'- r,_) O'J "-5
_l%<:i:[ _ (:]J
-_ .

o (lJ 8-e _ (]J (b C'_l(b _ E

_< o _ _ -_ __ _:_
--ElI.,9-C__ -U
_'9 (l] -- %o_
O _--I-_ -- _._ c-

c- Uo II I:D E:_
no ra c U (i_ _&_O c- "-1 c-
O _'_"_ tm _ :E_ > -- O ra _ -_-_-_ _9 (0 4-m "_ --

'_ I

_...1.._ _u_ c-_ _ E


.qD _ q._COw "_ _ I

_'_ '_ ...I.-_ _"- r,_) _ _ uO_X _ n_


_ v" U 0"_<:_ CO _- c- 0

,._- EE O _:_ I-- "_.._ .-_ "_

15,,t_ -- .i-_,J c- <12 u9 -- - (D-I_ c- _

u9 _ _-- _.,9qJ -- <:12f-- _: _.,9 _- (I.__-- I


_ c-.f_ ro c- E ,_ c-,_ ._ __ nu I
_ q._ na (1J._ c-"_:_ q--_ U<O I=: _ I
i

_ _ dJ uo _ u9 (b _- _o

O.J II _
_ ,--_ _ (3.) .-I.-_ 0 c- _ O
C- ',J U3 q..) ',..O (]..) _ ..i.-_ c"c- n ra u3 _ (3.)
_ O,,J (1._-.I._ Od _ qJ _3J _ _ .-_ .,_[_ _ (_p c- C',,J

_ 4-_
ra ra _(L_ ._4-_
L._ ra r..D
,<_ (L_ c-
c- ra '_O"_O_-
E c-
O UgEL
(]P _ C'U _(_J C- E_EE
(L _ "_ (_
-J c'-_ OJ .CO ._ _ _- 4-_-- U .._ X ra c- 4.-_ :>

_._ _ c- _._'_ .. C_ <_ _4J O n:_ c- (L_F-- <_ u _--

_'_ _._ <_ c- c- Or"'..-[_'._ _ _ N c- > __ _-_ uo _ ru __c- _-


E u'_ "_D O O _ . -(_ c".._ ra r_ r_ ...i._ _"_ ra c- c- U ra_-_
r_ -s q.) (]p c"- .r_ _.f_ _ O_ u9 O - c-" (_ _,') q.) Q...)
, _"_ -- (b ru+._ ru-- -- ._._ c- - _- _- (_J c- (]J :>b-_::_ E E - O c-4,-_
O _- q_ (1) -_D q_ O E:)-4._ O n u9 ..t-_ O q_'_'_ O O ra4--

._ ,--_,_<>_ _-o=_ < _ _-,-,_,_ <,_ ....


._ _U_ 4_CO _''-
mmn_ O
_ _- O
--_ _ _.J
(]J-I O EqJ O'_
c" r'_ r'_
f..._ E'_
r_ c'-_ ch Oh
c- CY-I_
-1:2_ _,o
C__ _/3
_2__ _U..._ "-1 _:=:"_
r_ c-
"_
c- _
_-_ =__ _'_'_='_
G._ E u9 (1)
_,_,-8_-'6_'
_ 4-- ra u0 c-4-
_5 ul---_
"- __-"_
C_4-- rU _f_
O_ (].J ._ _ O..J (-" - 0 (lP "_" n 4--_ ,"_ 0 G._ - 4-_ r_ (]J f,.- (]..}

7
The Superpave software uses these test results to estimate pavement performance
predictions for permanent deformation, fatigue cracking, and low-temperature cracking. The
reliability of these predictions is consistent with the designation of level 2 as the method of
choice for routine mix designs. The optimum asphalt content is determined from these
performance predictions.

The use of gyratory compaction and the Strategic Highway Research Program
(SHRP) shear test device is essential to successful level 2 mix designs. This design level
also introduces a test for tertiary creep to screen out early in the design process trial mixes
that may be susceptible to catastrophic rutting failures or unacceptable long-term permanent
deformation 2. In addition, this design level explicitly considers the effects of moisture
sensitivity and aging in selecting the final mix design.

Together, level 2 and level 1 mix designs should be suitable for 95 percent or more
of the mix design work conducted by a state agency in a typical year; in many states, these
design levels will be satisfactory for all mix designs.

1.2.2.3 Level 3 (High Traffic) Mix Design

The Superpave level 3 mix design is presented as a flow chart in figure 1.4 and is
described in detail in chapter 5 of this manual.

The level 3 mix design incorporates the selection of a design asphalt content with
the volumetric (level 1) design procedure. Candidate mixes prepared at the design asphalt
content and at a high and a low asphalt content bracketing the design value are subjected to
a series of performance-based tests selected to develop mix designs suitable for very heavy
traffic, severe climates, or any situation where only a minimal design risk is tolerable. Level
3 mix designs require considerably more time and a greater number of specimens than level
2 designs.

The Superpave software uses these test results to estimate pavement performance for
permanent deformation, fatigue cracking, and low-temperature cracking. The degree of
reliability of these estimates is consistent with the designation of level 3 as the method of
choice for mix designs where high performance is mandatory. The optimum asphalt content
is determined from these performance estimates.

The use of gyratory compaction and the SHRP shear test device is essential to
successful level 3 mix designs. This design level also uses a test for tertiary creep to screen
out early in the design process those trial mixes that may be susceptible to catastrophic

2 It is anticipated that the test for tertiary creep will not be needed if the compactibility
of the mix conforms to the Superpave guidelines.
rutting failures or unacceptable long-term permanent deformation. In addition, this design
level explicitly considers the effects of moisture sensitivity and aging in selecting the final
mix design.

The level 3 mix design also provides a battery of optional proof tests recommended
to confirm the results of the performance-based tests for paving projects where the
reliability of the mix design must be unquestioned.

1.3 The Superpave Performance-Based Test Methods for Paving Mixes

The Superpavemix design systemuses a compactionmethodthat simulatesfield


compaction, field-validated conditioning procedures, and, in levels 2 and 3, a set of
performance-based test methods for paving mixes. These allow the development of trial mix
designs and the characterization of their engineering capabilities through the measurement
of fundamental material properties. These methods and procedures, and the equipment
required for each, are presented in table 1.2.

Table 1.2. Superpave Mix Design Tests and Equipment

Method or Procedure Test Equipment Relevant SHRP or Other Test


Designation

Gyratory compaction SHRP gyratory compactor M-002

Moisture sensitivity Testing machine per AASHTO T AASHTO T 283


167 or environmental or M-006
conditioning system

Short and long-term aging Forced draft oven M-007

Frequency sweep at constant Shear test device M-003, P-005


height

Simple shear at constant height Shear test device M-003, P-005

Uniaxial strain Shear test device M-003, P-005

Volumetric (hydrostatic) Shear test device M-003, P-005

Repeated shear at constant stress Shear test device M-003, P-005


ratio

Repeated shear at constant height Shear test device M-003, P-005


(optional)

Indirect tensile creep Indirect tensile test device M-005

Indirect tensile strength Indirect tensile test device M-005

10
1.3.1 Laboratory Compaction

The SHRP gyratory compactor is an effective tool to simulate the field compaction
process and to ensure that engineering properties of laboratory compacted specimens are
equivalent to those of the in-place paving mix. This unit is capable of central laboratory and
field control operations. It permits real-time determination of specific gravity and air voids
content during compaction; this capability is required for the volumetric (level 1) mix
design discussed in chapter 3. The gyratory compactor produces a cylindrical, 150 mm
diameter test specimen of paving mix through a combination of vertical consolidation
pressure and gyratory kneading effort. It is capable of producing specimens up to 150 mm
in height. The paving mix is prepared in the laboratory or at a plant site and brought to the
proper compaction temperature. The loose mix is placed into a heated mold, and a vertical
pressure of 0.6 MPa is applied by means of a vertical ram. The mold is set for a 1.25+
0.02 angle of rotation, and the specimen is compacted at 30 revolutions per minute to a
predetermined number of gyrations based on anticipated traffic or other factors. At
completion, the mold is removed and the test specimen is extruded immediately.

NOTE: Agencies already owning another gyratory compactor such as a gyratory


testing machine (GTM) meeting the requirements of ASTM D 3387 may use it to
produce compacted specimens in lieu of the SHRP gyratory compactor. The GTM
model (or any other gyratory compactor) must be capable of: 1) maintaining a
fixed angle of gyration equal to 1.25+ 0.02; 2) operating at a speed of gyration
of 30.0 rpm; 3) applying a constant ram pressure of 0.60 MPa; and 4)
producing 150 ram-diameter specimens.
IIIIIIIIIIII I IIIIIIIIIII I I

1.3.2 Test Procedures for Moisture Sensitivity

Either AASHTO T 283, Resistance of Compacted Bituminous Mixture to Moisture


Induced Damage, or SHRP Method of Test M-006, Determining the Moisture Susceptibility
of Modified and Unmodified Hot Mix Asphalt with the Environmental Conditioning System,
is used in the Superpave mix design system to evaluate the moisture sensitivity of trial mix
designs.

Use of SHRP method M-006 requires an environmental conditioning system 3. This


is a modified triaxial test unit in which the dynamic resilient modulus of a cylindrical or
prismatic mixture specimen can be continually measured as moisture is forced through it.
Moisture susceptibility is characterized by the resilient modulus ratio of conditioned to
unconditioned specimens.

3Developed under SHRP contract A-003A.

11
1.3.3 Conditioning Procedure for Aging

Principal control on aging in the Superpave mix design system is through the
combined use of the rolling thin film oven test and the pressure aging vessel to measure the
long-term propensity of the asphalt binder to aging. Since laboratory mixtures are made
with unaged asphalt binders, the conditioning procedure for paving mixes must mimic both
plant and pavement aging in these laboratory mixtures.

In the short-term aging procedure for paving mixes, loose mix is placed in a tray
(immediately after mixing) to a uniform depth. The mix is held in a forced draft oven for 4
hours at 135C, after which the mix is brought to the appropriate compaction temperature
and the specimen compacted. This procedure simulates the aging that takes place during
HMA production and the pavement construction process.

In the optional long-term aging procedure for paving mixes, compacted specimens
are placed (prepared from loose mix which has undergone short-term aging) in a forced
draft oven at 85C. The time of exposure in the oven varies depending on the length of
pavement service that is simulated. The recommended exposure time is 2 days which is
equivalent to about 10 years of pavement service. Longer periods can be utilized at the
designer's discretion.

1.3.4 Performance Testing with the Shear Test Device

The ability of a paving mix to resist permanent deformation and fatigue cracking is
estimated through the use of the shear test device. The series of tests listed in table 1.2
provides the material properties that are used in the Superpave pavement performance
prediction models. The development over time of permanent deformation and fatigue
cracking is predicted for the trial mix design in a particular pavement under specific traffic
and environmental conditions.

The shear test device simulates the comparatively high shear stresses that exist near
the pavement surface at the edges of vehicle tires; these stresses lead to lateral and vertical
deformation. The shear test device has the ability to apply vertical and horizontal loads
simultaneously to the specimen in order to simulate both the compression and shear forces
applied to the pavement by loaded tires.

The primary components of the test device are a load frame, vertical and horizontal
actuators, environmental and confining pressure control systems, and a computer-controlled
test operation and data acquisition system.

The nonlinear viscous and elastic material behavior of the paving mix are measured
through a series of distinct tests conducted with the shear test device. These tests capture
critical aspects of this material behavior: dilatancy in shear; stiffening with increased
confining stress; and temperature and rate dependence. In addition, the accumulation of

12
permanent strain in the specimen under repetitive shear stress is determined. A brief
description of each shear test follows. The actual material properties determined from these
test results are discussed in chapter 6.

1.3.4.1 Nonlinear Elastic Behavior

Three tests are required to describe the nonlinear elastic response of the paving mix.

Simple Shear at Constant Height Test: A specimen (typically 150 mm in diameter


and 65 mm in height) is maintained at constant height while a shear load is applied
at 70 kPa/s. The load is applied very rapidly to ensure that only the elastic response
is measured (i.e., virtually no creep occurs in the specimen) and yet slowly enough
to avoid inertial effects. The shear stress causes a horizontal displacement, and the
axial load is varied to maintain the specimen at constant height.

Uniaxial Strain Test: A specimen encased in a rubber membrane is subjected to an


axial load applied at a rate of 70 kPa/s. Confining pressure is controlled
simultaneously by closed-loop feedback to maintain the specimen at constant
circumference.

Volumetric (or hydrostatic state of stress) Test: A specimen is completely


surrounded by a rubber membrane. A radial LVDT is placed around the
circumference of the specimen to monitor lateral deformation. A confining pressure
is applied to all specimen surfaces at a rate of 70 kPas, and the change in its
perimeter is recorded.

1.3.4.2 Viscoelastic Behavior

A single test determines the viscoelastic response of the paving mix.

Frequency Sweep at Constant Height: This test employs a specimen and loading
configuration identical to that of the Simple Shear at Constant Height test. A
sinusoidal shear strain of 1 x 10.4 is applied at frequencies of 10, 5, 2, 1, 0.5, 0.2,
0.1, 0.05, and 0.02 Hz. The vertical load on the specimen is continually varied to
maintain the specimen at constant height. Analysis of the test data provides the
phase angle (d_)and complex shear modulus (G*) of the paving mix.

1.3.4.3 Tertiary Creep

A single test determines the accumulation of tertiary creep damage in shear at a state
of stress similar to that induced in a pavement by application of a loaded tire.

13
Repeated Shear at Constant Stress Ratio: The specimen is subjected to the
application of a haversine shear pulse while the vertical load is continually adjusted
to maintain it in constant proportion to the shear (horizontal) load. Typically,
20,000 load applications are applied. The linearity of the log-log relationship
between the accumulated permanent strain and the number of load cycles is
evaluated.

1.3.4.4 Rutting Potential (Optional)

A single test rapidly estimates the potential for rutting induced in a paving mix by
the shearing action of loaded tires over the anticipated service life of the pavement. This
optional test is described in Appendix A of SHRP Standard Practice P-005.

Repeated Shear at Constant Height: The specimen is subjected to the application of a


haversine shear pulse. The specimen height is kept constant through the action of a
vertical (axial) load actuator, using as feedback the output of an LVDT that
measures the relative displacement between end caps mounted on the specimen.
Typically, the test is run for 5,000 load applications or until a shear strain of 5
percent is obtained. The number of ESALs necessary to induce a given rut depth is
empirically estimated from the relationship between the log of the permanent shear
strain and the log of the number of repetitive shear cycles.

1.3.4.5 Performance Testing with the Indirect Tensile Test Device

The ability of a paving mix to resist the development of fatigue cracking and low-
temperature cracking is estimated through the use of the indirect tensile test device. The
series of tests listed in table 1.2 provides the material properties that are used in the
Superpave pavement performance prediction models. The development over time of fatigue
cracking and low-temperature cracking is predicted for the trial mix design in a particular
pavement under specific traffic and environmental conditions.

The indirect tensile test device consists of a testing machine, environmental chamber,
and a control and data collection computer. (The capabilities of the testing machine are
same as those presented in section 5.1 and Note 2 of ASTM Standard Test Method D 4123,
Indirect Tension Test for Resilient Modulus of Bituminous Mixtures; this method and
equipment is required for the Long-Term Pavement Performance program.)

Two test methods, described in the following sections, are conducted at low and
intermediate temperatures with the indirect tensile test device in the Superpave mix design
system. Analysis of the test results yields the stiffness master curve; the slope of the
stiffness versus loading time relationship; and the tensile strength. The capability of the
paving mix to resist the development over time of cracking through the application of load
or temperature-induced stresses is estimated from these material properties.

14
Indirect Tensile Creep Test: A 150 mm diameter specimen is subjected to a
diametral preconditioning load to obtain uniform deformation. A fixed static load is
then applied for 1000 s, while maintaining constant temperature. The horizontal and
vertical deformations are monitored during the entire loading period across a gage
length of one-quarter of the specimen diameter.

Indirect Tensile Strength Test: A compressive load is applied along the diametral
axis of 150 mm diameter specimen at a controlled vertical deformation rate until
failure occurs.

1.4 Interaction with the Superpave Software

The Superpave specification, mix design, and support software are integrated in the
Superpave program. When used in conjunction with this manual, the Superpave software
guides the mix design process from beginning to end. It provides an orderly, self-contained
means for recording all test data and analysis results, performance prediction, and other
record keeping required for a complete mix design at levels 1, 2, or 3. The need for manual
computations is eliminated at every stage of the laboratory mix design and during field
control operations. The flow of the Superpave software mirrors that of the mix design
system shown in figure 1.1.

At the conclusion of the design, an individual computer file is produced for the
archives containing all essential information about the project design, from the initial
selection of materials to field control during HMA production and pavement construction.

Figure 1.5 shows the general structure of the Superpave software in schematic form.
The program core 4 contains

* the SHRP performance-based asphalt binder specification and the algorithms and
weather databases required to choose an appropriate binder performance grade for
the project;

material databases of pertinent test results on asphalt binders and aggregates;

the algorithms required to conduct the volumetric (level l) mix design; and

direction and control routines necessary for the orderly transfer of data from
computer files generated by test equipment, through algorithms for the analysis of
materials properties, to the performance prediction models.

4Developed under SHRP contract A-001.

15
Figure 1.5. Structure of the Superpave Specification, Mix Design and Support
Program

ASSOCIATED SOFTWARE
Lood performancebosedtests
Non-loadperformoncebosedtests

16
Several complete, self-contained programs are embedded in the Superpave software
and accessed automatically by its direction and control routines:

algorithms 5 that calculate fundamental material properties, used in performance


prediction models, from the results of the load-related and nonload-related
performance tests described in section 1.3;

performance prediction models 6 that estimate the development of permanent


deformation, fatigue cracking, and low-temperature cracking over the life of the
pavement from the material properties of the paving mix and environmental,
structural, and traffic loading factors; and

a version of the Federal Highway Administration's Environmental Effects Model,


tailored for use with the Superpave software, that generates pavement temperature
files used by the performance prediction models.

Finally, there are several programs that are associated with the Superpave software
but are not part of it. These programs control the operation of the load-related and nonload-
related performance tests, conduct preliminary analyses of the test data, and store the data
in computer files directly readable by the Superpave software.

A significant part of the Superpave software is devoted to user-defined databases of


information on asphalt binders, aggregates, mix designs, construction history (field control),
and environment (weather). Another portion deals with specifications for asphalt binders
and aggregates. A typical user has read-only access to the specifications and environmental
databases; however, these files may be changed at the agency or consensus level.

A complete description of the Superpave software and instructions for its use are
contained in The User's Manual for Superpave System Software.

1.5 Use of Modified Materials

Modified asphalt binders will be needed to supply the required levels of binder
performance at extreme temperature or traffic conditions. The use of modified paving
mixes, containing fibers or hydrated lime for example, may be warranted for certain
stringent conditions of pavement service.

5Developed under SHRP contract A-005.

6Developed under SHRP contract A-005.

17
In general, the Superpave mix design system does not require special tests to judge
the effect of modified materials. Specifications and test methods, because they are
performance-based, are identical for unmodified and modified materials and paving mixes.
The only exception is in evaluating the fatigue-cracking behavior of paving mixes (see
chapters 4 and 5 of this manual). At present, the test method used to characterize fatigue-
cracking behavior, a frequency sweep with the SHRP shear test device, is directly
applicable only to unmodified HMA.

When preventing or reducing fatigue cracking is the principal goal of the use of a
modified paving mix, flexural beam fatigue tests (chapter 5) must be directly employed to
estimate the change in fatigue behavior compared to an unmodified paving mix.

SHRP Standard Practice P-001 (AASHTO Standard Practice PP5), Laboratory


Evaluation of Modified Systems, provides a means to quantify the increased performance
capabilities of modified binders and mixes or to guide the selection of appropriate modifiers
to achieve specific performance levels.

1.6 Multilayer Designs in the Superpave System

The Superpave system and software provides a complete design and analysis system
for projects with multiple new layers, such as constructing a base, binder, and surface
course on a project. The Superpave material specifications account directly for differences
arising from the position of the layer in the pavement structure. For example, the Superpave
system will consider the variation in temperature at different depths of pavement, which
may allow selection of a different binder grade in the base course than in the surface
course.

The Superpave software is capable of sequentially accessing and carrying out


analyses with external performance test data from as many as two distinct mix designs,
each with as many as three trial asphalt contents. As discussed in section 1.7, the material
properties from each layer can be combined to obtain an estimate over time of the
pavement performance for the multilayer design.

In addition, the Superpave software provides the capability to use default design
values for the moduli of existing, underlying layers (including asphalt concrete and portland
cement concrete, granular bases, and subgrade) when calculating pavement performance
estimates for a new mix design. The input data include the type, classification and thickness
of up to eight layers, in addition to the new asphalt layer for which the Superpave mix
design is carried out.

18
1.7 Ground Rules for Use of the Superpave Mix Design System

The Superpave mix design system has certain ground rules which the user must
know in advance and keep in mind while producing designs.

1.7.1 Accounting for the Effects of Moisture Sensitivity

In itself, moisture sensitivity (or damage) is not a pavement distress mechanism, but
it does lead to or accelerate the occurrence of distress mechanisms such as raveling,
cracking, or rutting. Moisture sensitivity may be considered in a performance-based mix
design system in two ways.

In one approach, the moisture sensitivity of a trial mix design is tested directly, and
the test result is compared to a pass/fail criterion. If the trial mix design fails, remedial
action, such as introducing an antistrip agent or completely redesigning the mix, is
necessary.

In the second approach, the effect of moisture conditioning on the estimated


pavement performance of the trial mix design is determined. Testing and analysis of both
dry and conditioned specimens will provide an appraisal of the performance penalty
expected from accelerated distress development due to pavement moisture.

The Superpave mix design system employs the first approach. The moisture
sensitivity of a trial mix design is assessed by use of AASHTO Standard Method of Test T
283, Resistance of Compacted Bituminous Mixture to Moisture Induced Damage, or SHRP
Standard Method of Test M-006, Determining the Moisture Susceptibility of Modified and
Unmodified Hot Mix Asphalt with the Environmental Conditioning System, and compared to
a pass/fail criterion.

If the trial mix design fails the moisture sensitivity test, remedial action is required
before proceeding with the remainder of the mix design process. The performance
characteristics of a trial mix design that passes the moisture sensitivity test are evaluated
without further moisture conditioning.

Superpave Ground Rule for Moisture Sensitivity Moisture-sensitive mix designs are
screened out or remedied. The performance characteristics of the mix design are then
evaluated on the premise that moisture damage will not contribute to the long-term
development of pavement distress.
I I fl flflll IIII I

19
1.7.2 Accounting for the Effects of Aging

Like moisture sensitivity, aging of the asphalt binder is not a pavement distress
mechanism, but it does affect the rate of those mechanisms, particularly of low- temperature
cracking, fatigue cracking, and permanent deformation. Aging is accounted for in the
Superpave system through both pass/fail tests and conditioning procedures.

The asphalt binder must satisfy a pass/fail mass loss requirement in the rolling thin
film oven test (AASHTO Standard Method of Test T 240, Effect of Heat and Air on a
Moving Film of Asphalt (Rolling Thin Film Oven Test)). This test is intended to simulate
the effect of hot mix production on the asphalt binder. The asphalt binder residue
conditioned in the rolling thin film oven test and in the pressure aging vessel must satisfy
specific rheological requirements related to pavement performance. These requirements
assess the behavior of the asphalt binder at the time of pavement construction and after
long-term (5 years or more) exposure in the pavement.

Compacted specimens used in a Superpave mix design are prepared from loose mix
that has been conditioned in a forced draft oven to simulate the HMA production process
and several years of pavement service. Therefore, all estimates of pavement performance
obtained in a level 2 or level 3 Superpave mix design are calculated from material
properties similar to those found in an in-service pavement.

The Superpave system also provides an optional conditioning procedure that


simulates about 10 years of pavement service through the long-term oven aging of
compacted specimens. Use of this procedure permits an estimate of the effects of long-term
pavement service on predicted pavement performance, particularly the development of
fatigue cracking and low-temperature cracking.
I

Superpave Ground Rule for Aging The aging of pavements in service is simulated
for both the asphalt binder and the paving mix and is directly accounted for when
the performance characteristics of the mix design are estimated.
II I I I I I I I I I I II I I I IIII .....

1.7. 3 Superpave Mix Designs for New Construction

In order to facilitate the interaction between the Superpave mix design system and
the Superpave software, specific ground rules govern the treatment of the three pavement
distresses, permanent deformation, fatigue cracking, and low-temperature cracking, in level
2 or level 3 mix designs for new construction.

2O
For permanent deformation, a distinct mix design can be performed for each HMA
layer, and the Superpave software can use the computed performance characteristics of up
to two new layers to estimate the expected pavement performance over time. For projects
with three or more new HMA layers, selected layers must be combined in order for the
software to estimate pavement performance.

For fatigue cracking, the mixture characterization is performed for the lowest layer
only. For projects where several new HMA layers, each with a distinct mix design, are
being placed, the material properties of the mix design used in the layer that is placed
within the bottom one-quarter of the overall HMA thickness are used in calculating the
predicted fatigue cracking over time. However, the layer thickness is assumed to be the
combined thickness of all the new HMA layers, and the material properties are those
associated with the material in the bottom one-quarter of the total layer thickness.

For low-temperature cracking, only the uppermost layer is evaluated. For projects
where several new HMA layers, each with a distinct mix design, are being placed, the
material properties of the structural layer that is closest to the pavement surface are used in
calculating predicted low-temperature cracking over time. As with fatigue cracking, the
layer thickness is the combined thickness of all new HMA layers.

the measured material properties Of mix designs for the following number of layers:
Permanent deformation-upto two layers.
,, Fatigue cracking-one layer within the bottom one-quarter of the new

1.7. 4 Superpave Mix Designs for HMA Overlays

In order to facilitate the interaction between the Superpave mix design system and
the Superpave software, specific ground rules govern the treatment of the three pavement
distresses, permanent deformation, fatigue cracking, and low-temperature cracking, in level
2 or level 3 mix designs for HMA overlays over existing HMA or portland cement concrete
pavements.

For permanent deformation, a distinct mix design can be performed for each HMA
layer, and the Superpave software can use the computed performance characteristics of up
to two new layers can to estimate the expected pavement performance over time. For
projects with three or more new HMA layers, selected layers must be combined in order for

21
the software to estimate pavement performance. Thus, the treatment of HMA mix designs
for permanent deformation is the same, regardless of whether the mix is intended for new
construction or for an overlay over an existing flexible or rigid pavement.

For fatigue cracking, no testing or pavement performance prediction is performed,


since mixture fatigue properties of the existing HMA or PCC are not known. Generally,
fatigue cracking is not a predominant form of distress in HMA overlays.

For low-temperature cracking, no predictions are made due to the prevalence of


reflection cracking.

Superpave Ground Rule for HMA Overlays on Existing HMA: The Superpave system
level 2 and level 3 mix design methods are not recommended for the mix design of
any HMA layer less than 50 mm in thickness.

In predicting pavement performance for multi-layer overlays on existing flexible or


rigid pavements, the Superpave system can accommodate the measured material
properties of mix designs for the following number of layers:
* Permanent deformation-up to two layers.
* Fatigue cracking-not evaluated.
* Low-temperature cracking-not evaluated.

1.7. 5 Accounting for Reflection Cracking

By design, the Superpave system does not provide the capability to directly predict
or control the occurrence of reflection cracking in HMA overlays on existing flexible or
rigid pavements. This research area was specifically excluded from the scope of the SHRP
asphalt research program during its planning stages 7.

7See Strategic Highway Research Program Research Plans, Transportation Research


Board, Washington DC, 1986, page A- 13.

22
i.................
_i_ii
iii i!i_i_i!iiiiiiii
i i ........ !_i_ii!ii_i_ __iiiii_
....
Superpave Ground Rule for Reflection Cracking: The Superpave mix design system
is not directly applicable to the prediction or control of reflection cracking in HMA
overlays over existing flexible ....
or rigid pavements,
I

1.7.6 Units of Measurement in the Superpave System

The current trend in the United States is toward adopting the metric system of units in the
fields of manufacturing, engineering, and science. Accordingly, the Superpave system uses
units presented in the Intemational or SI System of Units (see ASTM Standard for Metric
Practice E 380) wherever possible.

Superpave Ground Rule for Units of Measurement: The Superpave mix design
system uses the SI system of units wherever possible,

1.7. 7 Pavement Structural Factors in Superpave

Both level 2 and level 3 Superpave mix designs employ pavement structural factors,
specifically layer thicknesses and moduli, in the computation of predicted distresses.
Consequently, the better the quality of the structural data, the more reliable the final
decision on the optimum mix design. If little or no information is available on the
underlying pavement structure, the benefit of using Superpave techniques may be
diminished.

!i!i_i_F ....... !i!_!_i_ ..... _ii_


_ iii_
Superpave Ground Rule for Pavement Structural Factors: The better the structural
information, the more benefit is derived from the use of the Superpave mix design
method.

23
2

Selecting Mixture Materials

The first, critical step in the Superpave mix design process is selecting materials
suited to the demands of traffic and environment expected over time at the paving project.

This chapter presents guidelines and requirements for selecting aggregate, asphalt
binder, and modifier on the basis of specific pavement performance criteria related to

the environment (climate);


the anticipated traffic volume over the service life of the pavement; and
the pavement structure.

It also presents a method for determining the chemical compatibility of the asphalt binder
and aggregate in the presence of moisture.

A mix design must balance the fulfillment of performance requirements against the
cost and availability of materials. The Superpave system simplifies this balancing process
by providing objective performance criteria that are used to directly assess the potential
benefits or penalties of the use of one material compared to another and decide on the most
cost-effective combination of materials that satisfies the project specifications.

2.1 Aggregate

The Superpave mix design system contains specific performance-related


characteristics (sections 2.1.1 through 2.1.9) to assist in selecting acceptable aggregate
materials and developing a satisfactory aggregate blend for a given situation. Aggregate
characteristics are specified as either agency standards or consensus standards. Except as
noted in section 2.1.1, the aggregate criteria in this section apply to materials intended for

25
dense graded paving mixes as well as special purpose paving mixes such as stone matrix
asphalt (SMA).

The Superpave mix design system accommodates coarse aggregate up to a maximum


size (see definition in section 2.1.1) of 50.0 mm. Table 2.1 presents recommended
maximum aggregate sizes for the design of paving mixes for base, binder, and surface
courses. This table provides guidelines for situations where the maximum size is not already
dictated by existing agency policy.

The SUPERPAVE" system provides three levels of mix design. This permits the
development of designs that will meet the demands of progressively higher levels of
traffic, environmental severity and reliability. Materials selection criteria are
recommended as either guidelines or requirements, depending upon the level of mix
design being employed. A guideline is defined as a specified value which is
provided as design guidance, but to which conformance is not mandatory. A
requirement is a specified value which the material must meet.

Table 2.1. Suggested Maximum Aggregate Sizes

PavementLayer NominalMaximumAggregateSize
Surface 9.5 - 12.5mm
Binder 25.0 - 37.5mm
Base 25 - 37.5 mm

Any sound, clean aggregate of local origin that meets these criteria may be chosen
as a candidate material in the Superpave mix design system. There are no requirements for
geologic type or composition other than the need to meet the optional asphalt binder
compatibility guideline in section 2.2.3 and the moisture sensitivity requirements for
mixtures presented in chapters 3, 4, and 5. If aggregates that meet the project requirements
are unavailable locally, the cost of importing suitable materials must be carefully weighed
against the penalties in long-term performance that will arise from the use of substandard
materials.

Techniques for gradation analysis and for the proportioning of two or more
aggregates to meet specific blended gradation requirements are beyond the scope of this
manual. Appendix A, "Gradation Analysis of Aggregates", of Asphalt Institute Manual
MS-2, Mix Design Methods for Asphalt Concrete, 1994 edition, gives complete, detailed
instructions on these topics.

26
The following rules apply generally to aggregate gradation analysis and the
development of aggregate blends in the Superpave mix design system:

Washed sieve analysis of all fractions, including filler, must be conducted in order to
ensure maximum accuracy in proportioning.

The gradation of an aggregate or aggregate blend is specified on the basis of the


total aggregate gradation, that is, on the total percent by weight passing the
designated sieve sizes.

Following the conventions stated in Asphalt Institute Manual MS-2, the individual
fractions of the total aggregate gradation are designated as shown in table 2.2.

Table 2.2. Definition of Aggregate Fractions

Coarse aggregate Retained on the 2.36 mm sieve

Fine aggregate Passing the 2.36 mm sieve

Mineral filler Passing the 75 lam sieve

When aggregate materials are designated by the terms rock, sand, and filler, these terms are
generally defined as shown in table 2.3:

Table 2.3. Definition of Aggregate Types

Rock Predominately coarse aggregate (retained on the 2.36 mm


sieve)

Sand Predominately fine aggregate (passing the 2.36 mm sieve)


Filler Predominately mineral dust (passing the 75 ktm sieve)

2.1.1 Gradation Control

The Superpave mix design system guides selection of an acceptable aggregate


gradation for a dense graded paving mix by means of control points and a restricted zone.
The control points and restricted zone are graphed on the Federal Highway Administration
grading chart on which the percentage of aggregate passing a sieve size is plotted against
the sieve opening size raised to the 0.45 power (figure 2.1). The ASTM sieves specified
for the SUPERPAVE system are presented in table 2.4.

27
A typical example of gradation control is shown in figure 2.2 for a 12.5 mm
nominal maximum aggregate size. This figure illustrates the following definitions:

Nominal maximum size: one sieve size larger than the first sieve to retain more than
10 percent of the aggregate.

Maximum size: one sieve size larger than the nominal maximum size.

Maximum density line: a line drawn from the origin of the 0.45 power chart to the
point at which the maximum sieve size intersects the 100 percent passing line.

Restricted zone: a zone lying on the maximum density line and extending from the
300 p.m sieve to the 2.36 mm sieve through which it is usually undesirable for the
gradation to pass. For 25 mm and 37.5 mm nominal maximum size gradations, the
restricted zone extends to the 4.75 mm sieve.

Control points: maximum and minimum limits established for each set of gradation
controls.

Table 2.4 illustrates the limits to which the blended aggregate gradation must
conform for a 12.5 mm nominal maximum size mixture.

Table 2.4. Aggregate Gradation Control Points

Control Point (Percent Passing)

Sieve Size Minimum Maximum

75 gm 2 10

2.36 mm 28 58

9.5mm -- 90

Nominal maximum (12.5 90 I00


mm)

Maximum (19.0 mm) 100 --

Tables A.1 through A.6 in appendix A present the control points and restricted zones for
gradations with nominal maximum sizes of 37.5 mm, 25.4 mm, 19.0 mm, 12.5 mm, and 9.5
him.

3O
Table 2.5 Common Sieves Sizes Used by State Highway Agencies

AASHTO Sieve Size and Approximate English Equivalent


Designation
63.0 mm 21Ain.

50.0 mm 2 in.

37.5 mm 11/zin.

25.0 mm 1 in.

19.0 mm in.

12.5 mm 1/zin.

9.5 mm 3/8in.

4.75 mm #4

2.36 mm #8

1.18 mm #16

600 _tm #30

300 _tm #50

150 _tm #100


75 lam #200

In general, it is recommended (but not required) that as the traffic level increases,
the aggregate gradation move toward the minimum control points, below the restricted
zone. Gradations that pass above or below the restricted zone, but within the relevant
control points, should produce acceptable mixtures in the Superpave mix design system.
No guideline or requirement is given for selecting the appropriate nominal maximum size.
Specifying agencies may select or specify a nominal maximum size for surface, binder, and
base course mixtures according to established policy, their past experience, or by reference
to the suggested nominal maximum sizes in Table 2.1.

Gradation control requirements in this section do not apply to special-purpose paving


mixes such as SMA or porous asphalt. Past experience and engineering judgment should be
used to develop specific gradation controls for the design and field control of special-
purpose paving mixes.

31
2.1.2 Coarse Aggregate Angularity

Coarse aggregate angularity is defined as the percent by weight of the aggregate


particles larger than 4.75 mm with one or more fractured faces. A fractured face is defined
as an angular, rough, or broken surface of an aggregate particle created by crushing, by
other artificial means, or by nature. A face is considered fractured only if it has a projected
area at least as large as one quarter of the maximum projected area (maximum cross-
sectional area) of the particle and also has sharp and well-defined edges _.

Coarse aggregate angularity is measured on the coarse particles of the blended


aggregate (specifically those retained on the 4.75 mm sieve) by Pennsylvania [Department
of Transportation] Test Method No. 621, Determining the Percentage of Crushed
Fragments in Gravel z. The results of this test are used in the laboratory during the
Superpave mix design process, as well as in field control to monitor aggregate production.

Coarse aggregate angularity criteria for increasing levels of total traffic in equivalent
single axle loads (ESALs) over the planned or estimated service life of the pavement are
presented in table 2.6. These are the requirements for level 1, 2, and 3 mix designs.

Table 2.6. Coarse Aggregate Angularity Criteria

Depth From Surface

Traffic (ESALs) < 100 mm > 100 mm

<3x105 55/- -/-

<lxl0 6 65/- -/-

<3x 106 75/- 50/-

<Ixl07 85/80 60/-

<3x107 95/90 80/75

<1 x 10g 100/100 95/90

>lxl0 g I00/100 100/100

Note: "85/80" denotes that 85 percent of the coarse aggregate has one fractured face and 80 percent has
two fractured faces.

1To check this criterion, hold the aggregate particle so that the face is viewed directly. If the
face constitutes at least 25 percent of the area of the outline of the aggregate particle visible at that
orientation, it is considered a fractured face.

ZSee appendix C of this report. ASTM is currently developing a standard test method for coarse
aggregate angularity. Pennsylvania DOT Method No. 621 is recommended for use until the ASTM
method is adopted.

32
2.1.3 Fine Aggregate Angularity

Fine aggregate angularity is defined as the percent of air voids present in loosely
compacted aggregate that passes the 2.36 mm sieve. Fine aggregate angularity is measured
on the fine aggregate portion of the blended aggregate by AASHTO Standard Method of
Test TP 33 (ASTM Standard Method of Test C1252), Uncompacted Void Content of Fine
Aggregate (as Influenced by Particle Shape, Surface Texture, and Grading). The results of
this test are used in the laboratory during the Superpave mix design process, but not as a
field control tool to monitor aggregate production.

Fine aggregate angularity criteria for increasing levels of total traffic in ESALs over
the planned or estimated service life of the pavement are presented in the table 2.7. These
are the requirements for level 1, 2, and 3 mix designs.

Table 2.7. Fine Aggregate Angularity Criteria

Depth From Surface

Traffic (ESALs) < 100 mm > 100 mm

<3x10 5 -- __

<lxl06 40 --

<3 x 106 40 40

<3 x 107 45 40

<1 x 10s 45 45

> 1x 108 45 45

Note: Criteria are presented as minimum percent air voids in loosely compacted fine aggregate.

2.1.4 Toughness

Aggregate toughness is defined as the percent loss of materials from the blended
aggregate during the Los Angeles Abrasion Test (AASHTO Standard Method of Test T 96,
Resistance to Abrasion of Small Size Coarse Aggregate by Use of the Los Angeles
Machine). The results of this test are used in the laboratory during the Superpave mix
design process and may be used as a source acceptance control for aggregate suppliers. No
specific levels of aggregate toughness are required in the Superpave mix design method.
Agencies should specify relevant guidelines or requirements in standards for their local
situation.

33
2.1.5 Soundness

Aggregate soundness is defined as the percent degradation of the blended aggregate


during the sodium or magnesium soundness test (AASHTO Standard Method of Test T
104, Soundness of Aggregate by Use of Sodium Sulfate or Magnesium Sulfate). The results
of this test are used in the laboratory during the Superpave mix design process and may be
used as a source acceptance control for aggregate suppliers.

No specific levels of aggregate soundness are required in the Superpave mix design
method. Agencies should specify relevant guidelines or requirements in agency standards
for their local situation.

2.1.6. Deleterious Materials

Deleterious materials are defined as the percent by weight of undesirable


contaminants, such as soft shale, coal, wood, or mica--in the blended aggregate. This is
measured with AASHTO Standard Method of Test T 112, Clay Lumps and Friable
Particles in Aggregate. The results of this test are used in the laboratory during the
Superpave mix design process and may be used as a source acceptance control for
aggregate suppliers. No specific levels for deleterious materials are required in the
Superpave mix design method. Agencies should specify relevant guidelines or requirements
in standards for their local situation.

2.1.7. Clay Content

Clay content is a measure of the amount of clay material present in the portion of
aggregate that passes the 4.75 mm sieve. Clay content is measured on the portion of the
blended aggregate passing the 4.75 mm sieve by means of the sand equivalent test
(AASHTO Standard Method of Test T 176, Plastic Fines in Graded Aggregates and Soils
by Use of the Sand Equivalent Test). The results of this test are used in the laboratory
during the Superpave mix design process, as well as in field control to monitor aggregate
production.

The sand equivalent (SE) of the aggregate fraction is defined in AASHTO T 176 by
the equation

SE = SR xlO0
CR

where SR = the sand reading from AASHTO T 176; and


CR = the clay reading from AASHTO T 176.

Therefore, larger values of SE are indicative of lower clay contents in the aggregate.

34
Clay content criteria for increasing levels of total traffic in ESALs over the planned
or estimated service life of the pavement are presented in table 2.8. These are requirements
for level 1, 2, and 3 mix designs.

Table 2.8. Clay Content Criteria

Traffic (ESALs) Sand Equivalent


<3x106 40
<3x107 45
>3x107 50

2. 1.8 Thin, Elongated Particles

The term thin, elongated particles denotes the coarse aggregate particles which have
a ratio of maximum to minimum dimensions greater than five (5).

The percentage of thin, elongated particles is measured on the portion of the blended
aggregate retained on the 4.75 mm sieve by ASTM Standard Method of Test D 4791, Flat
or Elongated Particles in Coarse Aggregate. The results of this test are used in the
laboratory during the Superpave mix design process and may be used as a source
acceptance control for aggregate suppliers.

Maximum criteria for thin, elongated particles for increasing levels of total traffic in
ESALs over the planned or estimated service life of the pavement are presented in table
2.9. These are requirements for level 1, 2, and 3 mix designs.

Table 2.9. Criteria For Thin, Elongated Particles

Traffic (ESALs) MaximumWeightPercentof Thin, Elongated


Particles
>lxl0 6 10

2.1.9. Dust Proportion

Dust proportion is defined as the ratio of the percent by weight of aggregate passing
the 75 _m sieve to the effective asphalt binder content expressed as percent by weight of
the total mix. Thus, the asphalt binder content does not include the asphalt binder absorbed
by the aggregate. Dust proportion is calculated and checked during the volumetric (level 1)
mix design as a measure of mix acceptability. Criteria for dust proportion for all traffic
levels over the planned or estimated service life of the pavement are presented in table 2.10.
These are suggested for level 1, 2 and 3 designs for dense graded paving mixes. It should

35
be noted, however, that these suggested limits are very dependent on the size and specific
gravity of the dust. Dust smaller than 20 to 30 lxm may act as an extender of the binder
rather than a filler. Fillers with low specific gravities will represent a greater volume of
dust.

Table 2.10. Criteria For Dust Proportion

Traffic(ESALs) Dust Proportion


All levels 0.6 - 1.2

2.2 Asphalt Binders

The performance-based specification for asphalt binders within the Superpave system
is designed to quantify and maximize the performance of the binder in reducing the
occurrence of permanent deformation, fatigue cracking, and low-temperature cracking.

2.2.1 General Treatment of the Asphalt Binder in the Superpave System

The choice of asphalt binder grade alone will not eliminate permanent deformation
which is strongly dependent upon the aggregate properties and the volumetric properties of
the as-constructed paving mix, or fatigue cracking which is also highly dependent upon
pavement structure. However, selecting the proper grade asphalt binder essentially
eliminate low-temperature cracking. The Superpave mix design system facilitates selecting
asphalt binders that provide different levels of protection or reliability.

The performance grade of an asphalt binder (PG x-y) is verified in accordance with
AASHTO Standard Practice PP6, Practice for Grading or Verifying the Performance Grade
of an Asphalt Binder. The following is a summary of the verification procedure:

(1) Determine the flash point of the asphalt binder in accordance with AASHTO T 48.
Report the flash point temperature (C).

(2) Test the binder with a rotational viscometer in accordance with ASTM D 4402 at
135C. Report the viscosity.

(3) Test the asphalt binder in accordance with AASHTO TP5 at Tm_x (generally 45C to
75C). Determine G*/sin 8. (Tmax is determined from the average 7-day maximum pavement
design temperature; see AASHTO specification MP1. G* is the complex shear modulus; 8
is the phase angle).

(4) Test the asphalt binder in accordance with AASHTO TP1. At 1 and 24 hours, determine
the creep stiffness and the slope of creep stiffness versus time relationship at Tmin -b 10C

36
(generally -36C to 0C). Calculate and report the physical hardening index, h. (Tmin is
determined from the minimum pavement design temperature; see AASHTO specification
MP1).

(5) Age the asphalt binder in the rolling thin film oven test (RTFOT) (AASHTO T 240).
Report the percent mass loss.

(6) Test the RTFOT residue of the asphalt binder in accordance with AASHTO TP5 at Tmax
(generally 45C to 75C). Determine G'/sin 8.

(7) Age the RTFOT residue in the pressure aging vessel (PAV) (AASHTO PP1).

(8) Test the RTFOT-PAV residue of the asphalt binder in accordance with AASHTO TP1.
Determine the creep stiffness and the slope of creep stiffness versus time relationship at
Tmi" + 10C (generally -36C to 0C).

(9) Test the RTFOT-PAV residue of the asphalt binder in accordance with AASHTO TP5
at Tavg(generally 7C to 34C). Determine G/*sin 8. (Tavgis a function of Tmax and Train; see
AASHTO specification MP1.)

(10) If necessary, test the RTFOT-PAV residue in accordance with AASHTO TP3 at Tmin +
10C (-36C to 0C). Determine the failure strain.

(11) Compare all test results with the specification requirements for the performance grade,
PG T,,_- T,,,,.

The Superpave asphalt binder specification (figure 2.3a and b) is stored in the
Superpave software for use as a look-up table to categorize asphalt binders according to test
results in SHRP Practice P-002. Additional grades can be added as needed. Test results
are entered into the software, and can be checked for conformance with the required
specification limits. During the mix design, the specification is used as a look-up table to
determine the grade of asphalt binder required for a specific project. This is based upon the
selection of high and low design pavement temperatures for the project (section 2.2.2), and
consideration of traffic.

2.2.2 Selecting an Asphalt Binder Performance Grade Based Upon High and
Low Pavement Design Temperatures and Traffic Level

The procedures in t_s section correspond with block (1,2) in figure li2, the flow

37
Figure 2.3a Performance Graded Asphalt Binder Specification (AASHTO MPI)

PG 46- PG 52- PG 58- PG 64-


PERFORMANCE GRADE
34 [ 40 [ 46 10 I16 [22 [ 28 134140 [ 46 16 [ 22 I 28 [ 34 [ 40 10 116 [ 22 I 28 I 34 I 40
Average 7-day Maximum Pavement Desig_
Temperature, C _ <46 <52 <58 <64

TeM'_ipmUatuPea:ecent Design 1>-341 >_0 >101 > + I> I> I>_t>_ >1 >1 >1 > 341>_ >1[>-1_ >-221 >-281 >-34[ >-40

ORIGINAL BINDER

Flash Point Temp, T48: Minimum C 230

Viscosity, ASTM D4402. _


Maximum, 3 Paos, Test Temp, C 135
Dynamic Shear, TP5: c
G'/sin_, Minimum, 1.00 kPa 46 52 58 64
Test Temp @ 10 rad/s, C

ROLLING THIN FILM OVEN

Mass Loss, Maximum, percent 1.00


Dynamic Shear, TP5: 64
G'/sin_, Minimum, 2.20 kPa 46 52 58
Test Temp @ 10 rad/s, C

PRESSURE AGING VESSEL RESIDUE (PP1)


PAV Aging Temperature, C d 90 90 100 100

Dynamic Shear, TP5:


G'sin_, Maximum, 5000 kPa
Test Temp @ 10 rad/s, C 10 7 4 25 22 19 16 13 10 7 25 22 19 16 13 31 28 25 22 19 16
Physical Hardening c Report
Creep Stiffness, TPI: r
S, Maximum, 300 MPa,
m - value, Minimum, 0.300 -24 -30 -36 0 -6 -12 -18 -24 -30 -36 -6 -12 -18 -24 -30 0 -6 -12 -18 -24 -30
Test Temp @ 60s, C
Direct Tension, TP3: t
Failure Strain, Minimum, 1.0% -24 -30 -36 0 -6 -12 -18 -24 -30 -36 -6 -12 -18 -24 -30 0 -6 -12 -18 -24 -30
Test Temp @ 1.0 mm/min, C

Pavement temperatures are estimated from air temperatures using an algorithm contained in the Superpave software program, may be provided by the
specifying agency, or by following the procedures as outlined in PPX.
h This requirement may be waived at the discretion of the specifying agency if the supplier warrants that the asphalt binder can be adequately pumped
and mixed at temperatures that meet all applicable safety standards.
c For quality control of unmodified asphalt cement production, measurement of the viscosity of the original asphalt cement may be substituted for
dynamic shear measurements of G'/sin/_ at test temperatures where the asphalt is a Newtonian fluid. Any suitable standard means of viscosity
measurement may be used, including capillary or rotational viscometry (AASHTO T201 or T202).
The PAV aging temperature is based on simulated climatic conditions and is one of three temperatures 9if'C, 100C or ll0C. The PAV aging
temperature is 100C for PG 58- and above, except in desert climates, where it is llOC.
e Physical Hardening -- TP1 is performed on a set of asphalt beams according to Section 13.1, except the conditioning time is extended to 24 hrs + 10
minutes at 10C above the minimum performance temperature. The 24-hour stiffness and m-value are reported for information purposes only.
t If the creep stiffness is below 300 MPa, the direct tension test is not required. If the creep stiffness is between 300 and 600 MPa the direct tension
failure strain requirement can be used in lieu of the creep stiffness requirement. The m-value requirement must be satisfied in both cases.

38
Figure 2.3b Performance Graded Asphalt Binder Specification (AASHTO MP1)
(Continued)

PG 70- PG 76- PG 82-

PERFORMANCE GRADE 10 116 122 12s 134 140 10 116 122 12s 134 10 [ 16 [ 22 [ 28 I 34
Average 7-day Maximum < 70 < 76 < 82
Pavement Design Temp, C b
Minimum Pavement
Yemperatnre'Cb
Design
>01> >01> >01
ORIGINAL BINDER
Flash Point Temp, T48: Minimum C 230

Viscosity, ASTM D4402P


Maximum, 3 Pa.s, Test Temp, C 135

Dynamic Shear, TP5: _


G'/sin6, Minimum, 1.00 kPa 70 76 82
Test Temp @ 10 rad/s, C

ROLLING THIN FILM OVEN ('1"240)


Mass Loss, Maximum, percent 1.00
Dynamic Shear, TP5:
G*/sin_, Minimum, 2.20 kPa 70 76 82
Test Temp @ 10 rad/s, C

PRESSURE AGING VESSEL RESIDUE (PP1)


PAV Aging Temperature, C d 100(110) 100(110) 100(110)

Dynamic Shear, TP5:


G'sin$, Maximum, 5000 kPa 34 31 28 25 22 19 37 34 31 28 25 40 37 34 31 28
Test Temp @ 10 rad/s, C
Physical Hardening* Report
Creep Stiffness, TPI: r
S, Maximum, 300.0 MPa,
m - value, Minimum, 0.300 0 -6 -12 -18 -24 -30 0 -6 o12 -18 -24 0 -6 -12 -18 -24
Test Temp @ 60s, C
Direct Tension, TP3: f
Failure Strain, Minimum, 1.0% 0 -6 -12 -18 -24 -30 0 -6 -12 -18 -24 0 -6 -12 -18 -24
Test Temp @ 1.0 mm/min, C

39
The correct performance grade of asphalt binder is determined in the Superpave
paving mix design process through consideration of the climate and the type of traffic
loading at the site of the paving project. The use of this performance grade is a suggested
requirement for level 1, 2, and 3 mix designs. The Superpave software guides the mix
designer through this process. The statistical distributions, expressed as the means and
standard deviations, of the yearly 7-day average maximum air temperature and the yearly
1-day minimum air temperature are available in the Superpave software for 5313 weather
stations in the United States and 1515 weather stations in Canada.

The Superpave software calculates the distribution of design pavement temperatures


from the air temperature data, and guides selection of the minimum required performance
grade of asphalt binder that will satisfy the conditions. These distributions may be viewed
along with the degrees of probable risk associated with the selection of any particular
design temperature. Thus, a binder performance grade may be selected for the project that
either minimizes the probable design risk for high or low temperature pavement
performance, or accepts some higher degree of probable risk when required by agency
policy for the class of highway, the cost, and other relevant factors.

Figure 2.4 illustrates the relationship between the mean and the standard deviation of
the pavement temperature distributions, and the probability that in a given year the actual
temperature will not deviate beyond a certain value. Specifically, there is an 50 percent
probability in any given year that the actual temperature will not deviate beyond the mean,
an 84 percent probability that it will not deviate beyond the mean plus one standard
deviation (1S), a 98 percent probability that it will not deviate beyond the mean plus two
standard deviations (2S), and a 99.9 percent probability that it will not deviate beyond the
mean plus three standard deviations (3S).

In practice, the mix designer will be concerned with the probabilities in a given year
that the actual 7-day average maximum pavement temperature will exceed the historical
mean of the 7-day average maximum pavement temperatures and (or) that the actual 1-day
minimum air temperature will fall below the historical mean of the 1-day minimum
pavement temperatures. Since the temperature distributions represented by figure 2.4 are
approximately symmetrical around the mean values, there are corresponding probabilities
(less than 50 percent) that the actual temperature will fall below the historical mean of the
7-day average maximum air temperatures or exceed the historical mean of the 1-day
minimum pavement temperatures. Generally, these situations will not be of practical
importance since binder grade selection is aimed at the prevention of distress development
in normal weather years or years in which the weather is more severe than normal.

Several examples may be presented to illustrate the use of these temperature


distributions in binder grade selection, and the interaction of the fixed 60C steps between
(high and low temperature) performance grades with the probabilities developed from the
means and standard deviations. In example 1 (figures 2.5a and 2.5b), the distribution of
7-day average maximum pavement temperatures at a hypothetical weather station has a

40
Figure 2.4. Typical Probability Distribution of the 7-Day Average Maximum
Pavement Temperature

% PROBABILITY
50% 84% 98%
!

2S 3 S_ v'-'

7-DAY AVERAGE MAXIMUM PAVEMENT TEMPERATURE

4].
Figures 2.5a and 2.5b. Example 1: Probabilities for a Mean of 57C and a
Standard Deviation of 2C

MEAN OF DISTRIBUTION, 57C

99.9%

57 61
,, 64
7-DAY AVERAGE MAXIMUM PAVEMENT TEMPERATURE

MEAN OF DISTRIBUTION, 57C


69% PROBABILITY

I I
57 58 61 63
7-DAY AVERAGE MAXIMUM PAVEMENT TEMPERATURE

42
mean of 57C and a standard deviation of 20C. The mean is one degree below the nearest
performance grade step. There is a 69 percent probability in a given year that the actual
temperature will not exceed 58C and a 99.9 percent probability that it will not exceed
64C. Selection of a PG 52 grade for this location would not provide an adequate level of
high temperature protection since the risk (defined here as 100 - Probability) in a given
year that the actual temperature would exceed 52C is considerably greater than 50 percent.
Selection of a PG 58 grade (the next step up from a PG 52) would reduce the risk to 31
percent (100-69) while going up one more step (to PG 64) would reduce the risk to
virtually zero.

Example 2 (figures 2.6a and 2.6b) illustrates the situation where the mean 7-day
average maximum pavement temperature is 590C and the standard deviation is 2C. The
mean is one degree above the nearest performance grade step. In this case, selection of a
PG 58 grade, although only one degree below the mean temperature, would entail a 69
percent risk (100-31) that the high temperature performance of the binder would be
inadequate in any given year. Selection of the next higher grade, a PG 64, would reduce the
risk to less than one percent (100-99.4)

Finally, example 3 (figure 2.7) illustrates the situation where the mean 7-day average
maximum pavement temperature of 61C and the standard deviation is 2C. The mean is
midway between two performance grade steps (PG 58 and PG 64). In this case, selection of
a PG 58 would entail a very high risk (greater than 93 percent) of inadequate high
temperature performance in any given year. Selection of a PG 64 grade would be almost
mandatory, and would reduce the risk to less than seven percent. The same type of analysis
would apply to the selection of the low-temperature performance grade on the basis of the
mean and standard deviation of the historical distribution of 1-day minimum pavement
temperatures.

Alternatively, the Superpave software will accept a performance grade of asphalt


binder and estimate the degree of risk engendered by its use at a particular project site. This
capability in the software provides a rapid means of evaluating the suitability of locally
available or low cost materials.

This procedure for performance grade selection assumes that the pavement will
experience an average mix of car and truck traffic moving at moderate to high speeds (the
fast transient condition). Figure 2.8 permits an upward adjustment of the maximum design
temperature-based performance grade to compensate for: 1) a larger than average
proportion of slow-moving, heavy trucks or a frequent incidence of heavy standing loads;
and 2) expected traffic volumes in excess of 107 ESALs.

43
Figures 2.6a and 2.6b. Example 2: Probabilities for a Mean of 59C and a
Standard Deviation of 2C

MEAN OF DISTRIBUTION, 59C

31% PROBABILITY

I I
55 58 59 6:5
7'-DAY AVERAGE MAXIMUM PAVEMENT TEMPERATURE

MEAN OF DISTRIBUTION, 59C

I_Y

I I I I I ,
54.5 59 65 64
7-DAY AVERAGE MAXIMUM PAVEMENT TEMPERATURE

44
Figure 2.7. Example 3: Probabilities for a Mean of 61C and a Standard Deviation
of 2C

MEAN OF DISTRIBUTION, 61C

6.7% _ .3%
PROB

57 58 61 64 65
7-DAY AVERAGE MAXIMUM PAVEMENT TEMPERATURE

45
For example, suppose that a PG 58-28 grade of asphalt binder is selected on the
basis of maximum and minimum design pavement temperatures for a pavement project
adjoining a ready-mix concrete plant. The pavement experiences a much higher than normal
volume of slow-moving, heavily loaded concrete trucks. Using figure 2.8, the binder grade
is moved up to PG 64-28 to enhance the ability of the binder to resist pavement rutting at
high temperatures. Thus, the effect of slow-moving, heavy traffic is compensated by a 6C
increase in the design high pavement temperature, equivalent to raising the high temperature
performance range of the binder by one grade.

Similarly, suppose that the asphalt binder will be used on a pavement experiencing
fast transient traffic with an expected volume of 3 x 10 7 ESALs. Using figure 2.8, the
binder grade is moved up to PG 64-28 to enhance the binder's ability to resist rutting at
high temperatures.

In summary, selecting a design asphalt binder grade requires the following steps
carried out with the aid of the Superpave software:

(1) Select weather stations in the vicinity of the paving project. Weather data from as many
as three stations may be evaluated to estimate the climate at a paving site remote from
established stations. (Note: At its discretion, the agency may input specific weather data for
the site.)

(2) Select a degree of design reliability for high and low temperature performance. The
reliability for a particular project is established by agency policy or assigned on the basis of
the engineer's judgment of direct and indirect costs for maintenance and rehabilitation.

(3) Estimate the design pavement temperatures corresponding to the assigned reliability at
the location of the paving project.

(4) Determine the minimum required performance grade of asphalt binder that will satisfy
the selected maximum and minimum design pavement temperatures (and the associated
risks).

(5) For paving projects in locations that experience, slow or heavy truck traffic,
frequent braking or acceleration of heavy vehicles, frequent, heavy standing loads, and (or)
traffic volumes above 107 ESALs, adjust the performance grade determined in step 4 using
figure 2.8.

At the discretion of the agency, asphalt binder performance grades may also be
determined from maps assigning specific grades to selected areas and classes of highway
within districts, counties, etc.

46
_1 o
I=, i_
II
"0 uJ

o
E
_ o_ o _ _ _ _-o
_, ,_ _o_p. _ 0_ -,,, _-'
o_ o_ _ -"
u.l_

dw_'
o _
_- ,,, m_" _z "'
" l.d
er'_
.J (._

_ _ _" 4- _

i_ 0-
t-_ i_=n-
bJbJ

0 '-_ 0- __ z uJ_

o
l,=
_ _E _-o 0
-- _P
_ cx] al) ,q- 0 _) 0._

.=. o p_p__ ,_ '_ ._-


o
w _ co co ,_-
o o _-
o zo
-w ,.,,,i
an-

_. ,-,
_
zzz-,l..IJIII.l--
-' _" _' _' _
I
-' -' .,_-_...
o- ._'_-
z_ A o o o o o _ _"3 ___z
,, o _ _ m q" o >__ mo_

Jl

_,- I_ o _n Do 3_n1_3cI_31 NOIS30 JN3_3A_d MO7


_: _ _:, ,,>- A A
vv _o0_1_

_ 47
2.2.3. Asphalt-Aggregate Compatibility with the Net Adsorption Test
(Optional)

The procedures in this section correspond with block (I,2A) in figure 1,2, the flow
chart for a level 1 mix design.

The net adsorption test (SHRP Test M-001, Measurement of Initial Asphalt
Adsorption and Desorption in the Presence of Moisture) gauges the ability of an asphalt
binder to adhere to the surface of the fine aggregate portion of the total aggregate. It also
assesses the strength of the initial adhesion of the asphalt binder to the aggregate in the
presence of moisture. This optional screening procedure evaluates the chemical
compatibility of an asphalt binder and fine aggregate fraction, as well as the effect of
moisture on this compatibility. It provides guidance on the stripping potential of an asphalt
binder-aggregate pair, but it should not be used as the sole determinant of the pair's
satisfactory resistance to moisture damage in a paving mix.

Measuring the net adsorption behavior of an asphalt binder-aggregate pair requires


the following steps (consult SHRP Method of Test M-001 for the full procedure):

(1) Prepare a solution of the asphalt binder in toluene (approx 1 g of asphalt binder per liter
of toluene).

(2) Remove 4 ml of the solution, dilute to a total volume of 25 ml in a volumetric flask,


and determine the light absorbance at 410 nm with a spectrophotometer that has been
zeroed with pure toluene.

(3) Place 50 g of the portion of the total aggregate passing the 4.75 mm sieve size and 140
ml of the asphalt binder solution in a 500 ml Erlenmeyer flask.

(4) Agitate the solution and aggregate on a shaker table for 6 hours.

(5) Remove a 4 ml sample of the solution from the flask, dilute to 25 ml with toluene, and
determine the light absorbance.

(6) Determine (from the difference in absorbance readings) the amount of asphalt that has
been adsorbed from the solution due to the chemical attraction of the molecular components
of the asphalt cement with the aggregate surface.

(7) Immediately after the second solution sample is taken, add 2 ml of water to the flask.

48
(8) Shake the solution for another 8 hours.

(9) Take a final 4 ml of solution from the flask at the end of 8 hours, dilute to 25 ml with
toluene, and determine the absorbance at 410 nm.

(10) Determine (from the increase in absorbance) the amount of asphalt binder that has
been displaced from the aggregate surface by water molecules.

(11) Report the net adsorption of the asphalt binder on the aggregate as the percent of
initially adsorbed asphalt binder remaining on the aggregate after the introduction of
moisture.

Criteria for net adsorption are presented in table 2.11. These criteria are suggested
guidelines for level 1, 2, and 3 designs for dense graded paving mixes.

Table 2.11. Suggested Net Adsorption Criteria

Net Adsorption(%) [ MoistureSensitivityof AsphaltBinder-Aggregate


Pair
I
>70 Acceptable
>55 Marginal
<55 Poor

Asphalt binder-aggregate pairs that exhibit acceptable behavior in this test must
always be tested for moisture sensitivity of compacted paving mixes in the level 1 design
method (block 1-8 of figure 1.2).

If an asphalt binder-aggregate pair exhibits marginal or poor behavior in the net


adsorption test, selection of asphalt binder or aggregate from different sources that meet the
criteria in sections 2.1 and 2.2.1 should be considered 3. If no other sources of acceptable
materials are available, the original pair should be used only if the results of the moisture
sensitivity test (block 1-8 of figure 1.2) on the compacted paving mix are satisfactory.

3The results of research conducted in SHRP contract A-003B and other SHRP studies
strongly suggest that the aggregate is the principal determinant of the moisture sensitivity of
paving mixes.
49
2.3 Modifiers

This section applies to modifiers intended either for direct addition to the asphalt
binder or for incorporation in the hot mix asphalt during production. The
SUPERPAVE" mix design system does not deal with modified aggregates.
I II

Modifiers can enhance the ability of paving mixes to resist the development of
pavement distress. There are no strict guidelines to determine beforehand when modifiers
may be required in a mix design or to select among the many modifiers intended to control
one distress or another. Nonetheless, it should be noted that when a binder grade is
selected that binder may have to be modified to satisfy the performance properties. The
modification technique is the responsibility of the supplier. In a sense, the need for
modification is determined by the initial binder grade selection. The need for a modifier
may also become evident at any one of several stages in the Superpave mix design process:

Past agency experience with similar materials or project requirements may


demonstrate the effectiveness of modifier use.

The Superpave system may indicate the need for a performance grade of asphalt
binder that can meet a very high or very low pavement design temperature with a
high degree of reliability (section 2.2.1). Generally, modified asphalt binders are
needed to meet such requirements.

In level 2 and 3 mix designs (chapters 4 and 5 of this manual), the estimated
pavement performance of the trial design may be unsatisfactory, requiring a change
in the trial design. Depending on the performance area of interest, modification
strategies such as use of a modified asphalt binder, addition of liquid antistrip agent,
treatment of the aggregate with hydrated lime, or even addition of fibers or fillers to
the mixture may be required to bring the estimated performance up to satisfactory
levels.

Once the need for modification is established, several factors complicate the
selection of a modifier to remedy a specific performance defect:

A wide array of materials is marketed commercially as modifiers for asphalt binders


and paving mixes.

The effectiveness of any modifier usually depends on the chemical and physical
properties of the asphalt cement and aggregate with which it is combined, as well as
on the modifier's concentration in the asphalt cement.

50
Introducing a modifier into a paving mix may significantly affect performance
characteristics other than the one targeted for enhancement.

Therefore, it is not possible to provide a straightforward list of modifiers that will remedy
specific pavement distress problems in all situations.

In the Superpave mix design system, AASHTO Practice PP5, The Laboratory
Evaluation of Modified Asphalt Systems, guides the selection of appropriate modifiers to
enhance the pavement performance of new mix designs. The practice employs the same test
methods to evaluate the performance characteristics of modified binders and mixes as those
used for unmodified materials. In addition, the practice contains several special methods
and procedures, such as a test for storage stability of modified asphalt binders, that may
facilitate the routine use of modifiers in mix design.

A general procedure for design and evaluation of a modified paving mix includes
the following steps (for details, refer to AASHTO Practice PP5).

(1) Identify the need to use a modifier on the basis of


a design grade of asphalt binder that exceeds the capabilities of available
unmodified asphalt cements;
a mix design that has unsatisfactory estimated performance;
agency policy; or
past experience with available materials or project requirements.

(2) Determine whether changes in the original materials--for example, using a different
trial gradation or increasing the level of angular coarse aggregate particles--will adequately
improve the estimated performance characteristics of the mix design.

(3) If changes in the initial materials or mix design are not sufficient, select a class of
modifier to address the specific performance characteristic, and consult with modifier
vendors or an agency database to choose representative materials for laboratory evaluation.

(4) Using the methods and procedures in chapters 3 through 5, evaluate the effect of
candidate modifiers on the performance-based properties of the asphalt binder and the
paving mix. Evaluate modifiers at concentrations recommended by vendors or as suggested
by past experience, with the goal of optimizing performance at an acceptable cost. Consider
the need to vary the performance grade of the base asphalt binder to obtain maximum
benefit from the modifier.

(5) Determine the safety, chemical compatibility, homogeneity (storage stability), and
workability of the successful candidate modified binders from step 4 using the methods and
procedures in AASHTO Practice PP5.

51
(6) Select the combination of modifier, base asphalt binder, and aggregate that best fulfills
the requirements of steps 4 and 5.

(7) Consult with the modifier vendor to determine any special safety, workability, or field
control requirements that must be observed during hot mix asphalt (HMA) production and
pavement construction.

2.4 Next Steps in the Design Process

At this point in the Superpave mix design procedure, the following steps are
complete:

* identification of the required performance grade of asphalt binder;

selection of aggregate stockpiles suitable for use in the volumetric (level 1) design;

an optional evaluation of the chemical compatibility of the asphalt binder-aggregate


pair in the presence of moisture; and

selection of a modifier as required by agency policy, past experience, or the results


of either a previous asphalt binder evaluation or a full level 2 or level 3 mix design.

In the next chapter, the design aggregate structure and design asphalt binder content
are determined through a volumetric (level 1) mix design process with gyratory-compacted
specimens.

52
3

Level 1 Mix Design (Low Traffic Levels Volumetric


Design)

3.1 Introduction

The Superpave level 1 mix design is based on empirical performance-related


aggregate and mixture properties. Empirical properties are used to ensure adequate
performance for lower volume pavements and to provide initial selection of a robust mix
design as a platform for level 2 and 3 mix design procedures (chapters 4 and 5,
respectively). Level 1 mix design is based on the proposition that performance properties
of the asphalt binder and aggregate characteristics such as crushed aggregate faces and
gradation, combined with volumetric properties such as air voids and voids in mineral
aggregate, can be used as surrogate mixture properties to ensure adequate performance.
Figure 3.1 shows the level 1 mix design flow chart.

Pavements with low traffic may not warrant the time and expense of testing for
performance-based properties associated with level 2 and 3 mix design procedures.
Therefore, volumetric design is recommended for low traffic projects having less than or
equal to 10 6 design equivalent single axle loads (ESALs). The value of 10 6 ESALs can be
changed at the discretion of the agency since it is purely a policy decision. Project traffic
is defined as the expected ESALs on the design lane during the design life of the pavement
structure. Design ESALs may be estimated using traffic information from any source or
may be calculated using average annual daily traffic (AADT), and Federal Highway
Administration vehicle-type and lane distribution factors.

3.2 Volumetric Properties

This section describes the volumetric properties of the paving mix that are used in the
level 1 design method as surrogate properties to ensure adequate performance.
Selection of an asphalt binder grade and aggregate with satisfactory characteristics is
described in chapter 2.

53
Figure 3.1. Level 1 SuperpaveMix Design(Note: All specimens are compacted
from pavingmix that has been short-term aged (SHRPM-007)).

Select aggregate based upon Select asphalt binder grades


specification criteria based upon high and low
gradation limits pavement design temperatures. (1-2)
- coarse aggregate
angularity I
- fine aggregate
angularity Conduct optional net adsorption
- clay content test to evaluate asphalt-aggregate .
- thin and elongated compatibility. (2 specimens) (1-2A).
particles (1-1) .......................................

J
Prepare three or four trial gradations
determined from control points
established on 0.45 power curve. (1-3)

Calculate an initial trial asphalt


content for each trial gradation
(Two specimens compacted to Ndesig-
per gradation = 6 or 8 specimens_ (1-4)

Evaluate compacted trial gradations select


a design aggregate structure calculate an
estimated design asphalt content. (1-5)

I
Determine moisture Compact specimens at 4 AC's around
susceptibility of design the estimated design asphalt content.
asphalt content and design (2 compacted specimens per AC = 8
gradation
(Compact 6using AASHTO
specimens at T7%
283 specimens) (1-6)
AV. Test three dry and
three wet.) (1-8)
I
I
Select the design AC that satisfies
Ithe volumetric criteria at N desin
and the Nmax and Ninit g

criteria. (1-7)

54
3.2.1 Definitions

Volumetric properties are defined in accordance with the representation of the volume of an
asphalt paving mix shown in figure 3.2.

Air voids (V_) are the percent by volume of air between coated aggregate particles in
the compacted asphalt mixture.

Voids in mineral aggregate (VMA) are the volume of compacted paving mix not
occupied by the aggregate when the volume of the aggregate is calculated from its
bulk specific gravity (not the effective or apparent specific gravities).

Absorbed asphalt volume (Vb3 is the volume of asphalt binder absorbed into the
aggregate. It is equal to the difference between the aggregate volume when
calculated with bulk specific gravity and with effective specific gravity. Absorbed
asphalt volume is represented by the overlap of asphalt volume and bulk aggregate
volume shown in figure 3.2.

Asphalt content (Pb) is the percent by weight of asphalt binder in the total mixture,
including asphalt binder and aggregate.

Effective asphalt volume (Vb_) is the volume of asphalt binder which is not absorbed
into the aggregate. It is represented by the portion of the asphalt binder volume
shown above the bulk aggregate volume in figure 3.2.

Voids filled with asphalt (VFA) is the percentage of VMA filled with asphalt binder.
It is the effective asphalt volume divided by the voids in the mineral aggregate.

A method for volumetric analysis of compacted paving mixes is presented in SHRP


Standard Practice P-004, Volumetric Analysis of Compacted Hot Mix Asphalt.

3.2.2 Air Voids

Air voids in compacted mixtures are defined as those air spaces contained between
the asphalt binder-coated aggregate particles. The percent of air voids in a compacted
paving mix is calculated from the bulk specific gravity of the compacted specimen
(AASHTO T 166) and the theoretical maximum specific gravity (AASHTO T 209) of the
paving mix by the equation

Va=100 Gram-Grab (3-I)


Gmm

55
Figure 3.2. Definitions of the Volumetric Properties of Compacted Asphalt Paving
Mix

AIR -_ AIR VOIDS (Va) VOIDS IIN

AGGREGATE(Vmo)
EFFECTIVE VOLUME
ASPHALT

AGGREGATE -
-

' L
OF ASPHALT (Vbe)

AGGREGATE
VOLUME
VOLUME
ABSORBED J

ASPHALT (Vba)

56
where

Va = air voids in the compacted specimen as a percent


of total volume;

Gram = maximum specific gravity of paving mixture; and

Grab = bulk specific gravity of compacted mixture.

The recommended design air voids for all levels of traffic is four percent (Table 3.1).

Table 3.1. Air Voids Criteria l

Traffic Designair voids


(ESALs) (%)
All Levels 4

3.2. 3 Voids in the Mineral Aggregate

The volume contained between aggregate particles in a compacted paving mix is


termed the voids in the mineral aggregate (VMA). It is composed of the air voids and the
effective asphalt volume. An adequate level of VMA ensures that sufficient asphalt binder
is provided for satisfactory durability. In a paving mix with inadequate asphalt content,
accelerated hardening of the asphalt binder occurs. This leads to raveling of the exposed
surface aggregate particles by abrasive traffic forces and to fatigue cracking because of an
inability of the pavement to flex under traffic. It may also contribute to moisture damage.

VMA may be determined by subtracting the volume of aggregate particles from the
volume of the compacted mixture. The volume of the aggregate particles is calculated
using the bulk specific gravity of the aggregate as determined by AASHTO T 84, T 85, and
T 100. The following equation is used to calculate VMA:

VMA = I00- GmbPs (3-2)


Gsb

1Specified values for Superpave specimens may not be directly comparable to existing mix
designs because oven curing at or near compaction temperature is required to allow absorption
and aging to occur. Therefore, the same aggregate gradation and asphalt binder content may
produce Superpave specimens with air voids different from Marshall or Hveem specimens.

57
where
VMA = voids in mineral aggregate (percent of bulk volume);

Gsb = bulk specific gravity of aggregate;

Gmb = bulk specific gravity of compacted paving mix (AASHTO T


166); and

Ps = aggregate, percent of total weight of mixture.

VMA is used to control the durability of paving mixes at low traffic volumes.
VMA criteria vary with the nominal maximum size of the aggregate gradation. Acceptable
values are shown in table 3.2 for mixes with four percent air voids. Recommended
minimum values for three and five percent air voids are shown in figure 3.3.

Table 3.2. Voids in Mineral Aggregate Criteria


Nominal Maximum Size Minimum Voids in
Mineral Aggregate (%)

9.5 mm 15.0

12.5 mm 14.0

19.0 mm 13.0

25.0 mm 12.0

37.5 mm 11.0

50.0 mm 10.5

58
Figure 3.3. Minimum Percent Voids in Mineral Aggregate (VMA) for 3, 4 and 5
Percent Air Voids

59
3.2. 4 Voids Filled with Asphalt

The percentage of voids in the mineral aggregate filled with asphalt is termed the voids
filled with asphalt (VFA). VFA is the volume of effective asphalt binder expressed as a
percentage of the VMA. The following equation is used to calculate the VFA:

VFA= V_A- Va x 10 O (3-3)


VMA

Experience has shown that VFA should lie within limits which prevent both mix instability
under high shear stress and accelerated aging. For low traffic volumes, the shear resistance
and aging properties are not measured directly; therefore, VFA is a criterion which must be
met in the volumetric design. Acceptable values of VFA for increasing traffic are shown in
table 3.3. Specification of VFA in addition to air voids and VMA effectively places a
maximum value on VMA and allows for greater asphalt binder contents for low traffic
volume conditions.

Table 3.3. Voids Filled With Asphalt Criteria

Traffic Level Design Voids Filled


(ESALs) With Asphalt
(%)
<3 x 105 70- 80

<3 10 6 65- 78

<1 x 108 65-75

> 1 x 108 65 -75

3.3 Outline of the Level 1 (Volumetric) Design Method

The Superpave level 1 (volumetric) design method includes three principal parts: (1)
selection of materials; (2) selection of a design aggregate structure (expressed as an
aggregate gradation); and (3) selection of a design asphalt (binder) content.

In practice, paving mixes containing differing aggregate structures and asphalt binder
contents are compacted with the gyratory compactor. The volumetric properties of the
paving mixes are evaluated at several points along the compaction-density curve in order to
select a satisfactory design aggregate structure and design asphalt content. The compaction
levels at which the evaluation takes place are determined by the traffic projected for the
pavement over its service life. Therefore, asphalt binder properties, aggregate
characteristics, and volumetric properties are used as surrogate values to ensure adequate
pavement performance.

60
Within each principal part of the level 1 design method, several specific steps are
required.

(1) Select materials.

Select the asphalt binder and aggregate stockpiles that meet the environmental
and traffic requirements of the paving project (chapter 2 of this manual).

Determine the bulk specific gravity of all aggregates proposed for blending
and the specific gravity of the asphalt binder.

(2) Select the design aggregate structure.

Blend trial aggregate gradations (preferably three or more) from selected


aggregate stockpiles.

Calculate an initial trial asphalt content and compact two specimens for each
trial gradation.

Select a design aggregate structure and a corresponding estimated design


asphalt content on the basis of satisfactory conformance of a trial gradation
with requirements for Va, VMA, and VFA at the initial ('Ninit), design (Ndesig_),
and maximum (Nm=) compaction levels (measured in terms of gyrations
applied with the SHRP gyratory compactor).

(3) Select the design asphalt (binder) content.

Compact two specimens at the estimated design asphalt content and at the
estimated design asphalt content + 0.5 percent and + 1.0 percent.

Select the design asphalt content on the basis of satisfactory conformance


with requirements for Va, VMA, and VFA at the initial (Ni,it), design (Ndesign)
and maximum (Nm_x)compaction levels.

Evaluate the moisture sensitivity of the design aggregate structure at the


design asphalt content and an air voids content (Va) of 7 percent.

3.4 Equipment

The equipment required for the preparation of compacted specimens is as follows:

Pan: metal, flat bottom, for heating aggregates.

Pans: metal, round, approximately 4-liter capacity, for mixing asphalt and

61
aggregates.

Oven and hot plate: electric, for heating aggregates, asphalt, and equipment as
required. Oven must also be capable of maintaining temperature for short-term
aging.

Scoop: for batching aggregates.

Containers: gill-type tins, beakers, pouring pots, or sauce pans, for heating asphalt.

Thermometers: armored, glass, or dial-type with metal stem, 10C to 230C, for
determining temperature of aggregates, asphalt, and asphalt mixtures, preferably
ASTM certified.

Balance: 5-kg capacity, sensitive to 1 g for weighing aggregates and asphalt.


Balance, 2-kg capacity, sensitive to 0.1 g for weighing compacted specimens.

Mixing spoon: large, or trowel: small.

Spatula: large.

Mechanical mixer: (optional) commercial bread dough mixer 4 1 capacity or larger,


equipped with two metal mixing bowls and two wire stirrers, or suitable equivalent.

Gyratory compactor: capable of providing a consolidation pressure of 0.6 MPa, an


angle of gyration of 1.25 degrees and speed of gyration of 30 rpm; SHRP model or
equivalent.

Cylindrical molds: large enough to accommodate the following specimen size


requirements: 150 mm diameter, 150 mm maximum height, and 90 mm minimum
height.

Extrusion jack or arbor press: for extruding compacted specimens from mold.

Gloves, welders: for handling hot equipment. Gloves, rubber: for removing
specimens from water bath.

Marking crayons: for identifying test specimens.

62
3.5 Gyratory Compaction

The Superpave level 1 (volumetric) mix design method requires specimen


compaction with a gyratory compactor in accordance with SHRP Standard Method of Test
M-002. This compactor can monitor the increase in specimen density (expressed as a
percent of its theoretical maximum specific gravity) with increasing compactive effort.
Superpave software analyzes the densification data directly from the SHRP gyratory
compactor; data can also be collected manually or with a dedicated computer connected to
the compactor. Compaction is carried out at an equiviscous temperature determined in
SHRP Method of Test M-002 or at an appropriate temperature selected by the mix designer.

Density is evaluated at three points along the densification curve as shown in figure
3.4. The design aggregate structure and design asphalt binder content are selected in the
level 1 (volumetric) mix design to produce a densification curve which passes through 96%
of theoretical maximum specific gravity at the design number of gyrations (N0esign);thus, the
design asphalt content is selected at 4% air voids at Ndesign. The value of Noesign used in a
mix design is selected from table 3.4; it is determined from the traffic level expected on the
pavement and the design 7-day maximum air temperature for the pavement site.

Table 3.4. Number of Initial (N), Design (Nd) and Maximum (Nm) Gyrations
Required For Various Traffic Levels and Maximum Temperature
Environments

Traffic Design 7-day Maximum Air Temperature (C)


(ESALs)
<39 39-41 41 -43 43-45

Ni Nd N m Ni N d Nm Ni Nd Nm Ni Nd Nm

<3 105 7 68 104 7 74 114 7 78 121 7 82 127

<1 x 10 6 7 76 117 7 83 129 7 88 138 8 93 146

<3 x 106 7 86 134 8 95 150 8 100 158 8 105 167

<1 x 10 7 8 96 152 8 106 169 8 113 181 9 119 192

<3 10 7 8 109 174 9 121 195 9 128 208 9 135 220

<1 x 108 9 126 204 9 139 228 9 146 240 10 153 253

>1 108 9 143 235 10 158 262 10 165 275 10 172 288

63
Figure 3.4. Typical Densification Curve Obtained With SHRP Gyratory Compactor

I00.0
(j 98.0
96.0
LI,J ,,,'t'r
94.0
0 ,h '_
w >- 92D /h
"r
_ I--
_ 90.0 , _, _4r, D
x Z 88,0 r _--
86.0
I,- 84.0
zuJ 82.0
on- 80.0 NtNrnAL NDESJeN NMAX
LU 78.0
n _r
76.0 = I j
I0 I00 150 I000

NUMBER OF GYRATIONS

64
At the maximum number of gyrations (Nm_, the paving mix must attain less than
98% of theoretical maximum specific gravity or an air voids content of more than two
percent. The value of Nm_x used in a mix design is presented in table 3.4 and determined
from Ndesign by the equation
10gNma x =i . l 0 .ZOgNdesig n (3 - 4 )

At the initial number of gyrations ('Ninit), the paving mix must attain 89% of
theoretical maximum specific gravity or less. The value of Ninit used in a mix design is
presented in table 3.4 and determined from Ndes_g n by the equation
1 ogNin i t = 0 . 4 510gNdesig n (3 - 5 )

These compaction level-density requirements are summarized in table 3.5.

Table 3.5. General Superpave Compaction Requirements

CompactionLevel RequiredDensity
(% of Theoretical Maximum Specific Gravity)

Ninit Cinit < 89

Ndesign Cdesig n = 96

Nma x Cma x < 98

65
EXAMPLE: What are the values of N_ait , Ndo_,g,,and Nmaxrequired for a Superpave
level 1 mix design conducted for a paving project with an anticipated traffic level of
8x 105 ESALs and an average design air temperature of 39C?

Referring to table 3.4, the value of Nd_si_,is found at the intersection of the row
labelled <1 xlO 6 and the column headed 39 - 41. Therefore, Nd_ign = 83.

Then,
log Ni,it = 0.451og(83) = 0.86
Ni.it = 7

and
log Nm_x = t.101og(83) = 2.11
Nm_ = 129

I III IIII II I I I -

The density calculated at any point in the compaction process from the weight of the
specimen and its height is termed the uncorrected density (Cux). The uncorrected density at
x gyrations is calculated as a percentage of the theoretical maximum specific gravity of the
paving mix by the equation:

w=
Cux=lO0 [ d--_] /Gmm (3-6)

where Vmx, the volume of the paving mix at x gyrations in mm3, is given by the equation

_d2hx (3 -7)
Vmx- 4

and Wmx = the weight of the paving mix at x gyrations in g;

Gram = the theoretical maximum specific gravity of the paving mix,

d = the inside diameter of the specimen mold in mm;

hx = the height of the specimen in mm; and

dw = density of water at the temperature at which Gram is measured, in g/mm.

66
At the completion of the compaction process, the bulk specific gravity of the
compacted specimen is measured using AASHTO T 166. The bulk specific gravity is used
to correct the uncorrected density C,x by the equation

CuxG_Vmm (3 -8)
Cx_ W_xdw

where C = corrected density expressed as a percentage of the theoretical maximum


specific gravity;

Vmm = the volume of the mixture calculated at the maximum number of gyrations;

Wmx= the weight of the paving mix at x gyrations in g;

Grab = the measured bulk specific gravity of the compacted paving mix at Nmax; and

dw = density of water at the temperature at which Gram is measured, in g/mm.

The corrected density is plotted against the logarithm of the number of gyrations to
produce densification curves. Figure 3.5 illustrates typical densification curves, showing the
change in corrected density (Cx) as the compaction progresses from five to a maximum of
230 gyrations. Four densification curves are presented, one for each of four asphalt contents
with the same aggregate structure. Note that as the asphalt content increases, the
densification curves shift to the left but remain roughly parallel.

3.6 Short-Term Aging of Selected Target Asphalt-Aggregate Blend

All specimens of paving mix compacted in the level 1 mix design are conditioned as
loose mix according to the short-term aging procedure in SHRP Standard Method of Test
M-007. This conditioning simulates the aging of hot mix asphalt (HMA) paving mixes
during field plant mixing operations. It also permits asphalt absorption to proceed to
completion.

3.7 Selecting the Design Aggregate Structure (Skeleton)

The purpose of this part of the level 1 mix design is to evaluate the effect of
aggregate structure on mixture volumetric properties, predominately VMA. The objective is
to select a design aggregate structure (skeleton) that meets the Superpave requirements.
Specifically, the design aggregate structure must:

provide adequate VMA at the design number of gyrations and 4% air voids;
meet density requirements at Ninit gyrations; and
meet density requirements Nmax gyrations.

67
I

In this section, the design aggregate structure is selected.

At this point in the level 1 design process, both asphalt binder and aggregate fractions
that meet the requirements for the paving mix as specified in chapter 2 have been
identified.

Several trial aggregate gradations will be developed and evaluated. An initial trial
asphalt content is calculated for each. Compacted specimens are prepared for each trial
gradation-trial asphalt content combination; their densification curves and volumetric
properties are evaluated; and a design aggregate structure is selected.
........... III I I IIII IIII

Selecting a design aggregate skeleton requires the following steps:

(1) Select aggregates meeting Superpave criteria for acceptable aggregate characteristics
(chapter 2 of this manual).
(2) Blend aggregate stockpiles to meet specified gradation controls. Typically, three
trial gradations or more covering the range of allowable gradations should be
evaluated (section 3.7.1).
(3) Calculate an initial trial asphalt content for each trial gradation (section 3.7.2).
(4) Compact specimens at the initial trial asphalt content (section 3.7.3).
(5) Evaluate paving mix densification curves for each trial gradation (section 3.7.4).
(6) Select a design aggregate structure which best meets the volumetric and densification
requirements (section 3.7.5).

3. 7.1 Aggregate Trial Blend Gradations

:2

The procedures in this section correspond with block (1_3) in figure 3.I, the flow chart
for a level 1 mix design.
I III I

68
Figure 3.5. Gyratory Compaction Densification Curves for Four Asphalt Binder
Contents

100

96
_>. "
.j_.. 94

90
wc._

- 86 4.3/o AC
84 4.8% AC
82 5.3/,, AC
8o
w o 5.8% AC
a. 78
>_ 76 i i J
0
I I0 I00 I000
NUMBER OF GYRATIONS

69
(1) Refer to section 2.1.1 for gradation control requirements. The main purpose of the
gradation evaluation is to determine whether a selected gradation which meets the gradation
controls is likely to meet the VMA requirements. VMA is influenced predominantly by
aggregate characteristics such as gradation, angularity, and surface texture.

(2) Using stockpiles selected in chapter 2, perform a blending analysis to determine which
gradation(s) will be used in the volumetric proportioning phase. The goal is to evaluate how
various volumetric proportions of the available stockpiles fit the gradation controls. See
chapter 3, Evaluation of Aggregate Gradation, Asphalt Institute Manual MS-2, Mix Design
Methods for Asphalt Concrete and other Hot Mix Types, 6th Edition, 1994, for a detailed
exposition of this procedure.

(3) Develop three or four trial aggregate gradations, and plot the results of the blending
analysis on a 0.45 power gradation analysis chart as shown in figure 3.6 confirming that
each meets the Superpave gradation controls (Appendix A).

Blends 1, 2, and 3 in figure 3.6 are three examples of trial aggregate gradations.

3. 7.2 Calculate an Initial Trial Asphalt Content for each Trial Aggregate
Gradation.

The procedure in this section and section 3.7.3 corresponds with block (1-4) in figure
3.1, the flow chart for a level 1 mix design.

70
Figure 3.6. Evaluation of the Aggregate Gradations of Three Trial Blends

19.0mm NOMINAL MAXIMUM SIZE MIXTURE


I00.0
90.0 "'""
80.0

70.0 TRIAL BLEND


60.0
50.0 .,. ....
_
z 40.0
30.0
20.0
I0.0

0.0
.075 236 12.5 19.0 25.0
SIEVE SIZE (RAISED TO 0.45 POWER),mm

71
The next step in the level 1 (volumetric) mix design is to estimate an initial trial
asphalt content Phi for each trial gradation. This is the starting point for the subsequent
laboratory determination of the level 1 design aggregate structure and design asphalt
content. The procedure in this section permits the calculation of an initial trial asphalt
content for any trial aggregate gradation. The actual calculations are performed by the
Superpave software. It provides a reasonably precise estimate of the asphalt content
required by the trial gradation to meet minimum VMA requirements (Table 3.2) on the
basis of aggregate characteristics, including specific gravity and nominal maximum size.

The procedure calculates an initial trial asphalt content that satisfies a minimum
criterion for allowable effective asphalt content while compensating for asphalt absorption
by the aggregate. It is recommended that bag house fines be added to simulate field
conditions if a bag house is to be used. The amount can be based on experience or a value
of two percent by weight of the aggregate can be used. The minimum effective asphalt
content is determined from a regression equation which relates it to the nominal maximum
sieve size of the trial aggregate gradation. Asphalt absorption is approximated from the bulk
and apparent specific gravities of the aggregate fractions used in the trial aggregate
gradation.

Calculating an initial trial asphalt content requires the following steps:

(1) Determine the bulk and apparent specific gravity of each aggregate fraction in the
trial gradation using AASHTO Standard Methods of Test T 84, Specific Gravity and
Absorption of Fine Aggregate, T 85, Specific Gravity and Absorption of Coarse
Aggregate, and T 100, Specific Gravity of Soils.

(2) Calculate the bulk and apparent specific gravities G of the total aggregate in the trial
gradation as follows:

G- (PI+Pz+ """+Pn)
PI
--+
P2 +...+
Pn (3 - 9 )

Gn

where P1, P2, .... P. = the weight percent of aggregate fraction 1, 2, ..., n in the total
aggregate, and

GI, G2,....G . = the bulk or apparent specific gravities of aggregate


fractions 1, 2, ..., n.

72
(3) Estimate the effective specific gravity 2 of the total aggregate in the trial aggregate
gradation by the equation
Gse=Gsb+O .8 (Gsa-Gsb) (3-I0)

where Gs_ = the effective specific gravity of the total aggregate,


Gsb = the bulk specific gravity of the total aggregate, and
Gsa = the apparent specific gravity of the total aggregate.

(4) Estimate the percent volume of asphalt absorbed into the aggregate (Vba) as follows:

vba=ws
(i---
-i-!-
) (3-11)
Gsb Gse

where Ws, the weight percent of the aggregate, is calculated as

P sX (l - Va )

__Pb+ Ps (3-12)
Gb Gse

where Pb = weight percent of asphalt binder, in decimal equivalent, assumed to be 0.05;


Ps = weight percent of aggregate, in decimal equivalent, assumed to be 0.95;
Gb = the specific gravity of the asphalt binder, measured or assumed to be 1.02;
and
Va = the volume of air voids, fixed at 4 percent.

Note that this procedure calculates the volume of asphalt absorbed into the aggregate and,
subsequently, _e initial _iat asphalt content at air voids content of fo_ percenL _e
design criterion present_ in table 3.ti .....

2 The Superpave mix design system includes a short-term aging step before the
compaction of all specimen; this generally permits asphalt absorption to proceed to completion.
Therefore, the effective specific gravity of Superpave - designed paving mixes will tend to be
close to the apparent specific gravity, in contract to other design methods where the effective
specific gravity generally will lie near the midpoint between the bulk and apparent specific
gravities.

73
(5) Estimate the volume percent of effective asphalt binder, Vbe , from the following
empirical regression equation: 3

Vbe=O. 176- (0. 0675) log (S n) (3-13)

where Vbe = the volume of effective asphalt binder content; and


Sn = the nominal maximum sieve size of the total aggregate in the trial aggregate
gradation, mm.

(6) Using the volume of absorbed asphalt binder (Vba) and the volume of effective
asphalt binder (Vbe), calculate the initial trial asphalt content (Pbi), expressed as
percent of asphalt binder by weight of the total paving mix, by the equation

Gb(Vbe+Vba) (3_14)
Pbi- (Gb(Vbe+Vba)
) +Ws

where G b= the specific gravity of the asphalt binder, measured or assumed to be


1.02; and
Ws = the weight of the aggregate.

The estimation of the initial trial asphalt content carried out above is based on

3 This regression equation is derived from an empirical relationship between VMA and Vbo
when the air voids content Va is equal to 4 percent:

Vbe=VMA-V_=VMA-4

and the relationship between VMA and nominal maximum sieve size of the aggregate in table
3.2.

74
an assumed asphalt content in step 4;
an assumed specific gravity of the asphalt binder in steps 5 and 6; and
an estimated effective specific gravity for the total aggregate in step 3.

M this point in the selection of the design aggregate structure, several trial aggregate
gradations have been developed andan initial trial asphalt content has been calculated for
each. In the following steps, actual specimens will be compacted for each trial gradation
at its initial trial asphalt content, and their volumetric properties will be evaluated in
order to select a satisfactory design ag_egate structure.

The procedure in this section corresponds to block (1-5) in Figure 3.1, the flow chart for
Level 1 mix design.

3.7. 3 Compact Specimens of Each Trial Gradation

Duplicate specimens are mixed and compacted at the initial trial asphalt content for
each of the chosen trial aggregate gradations. The compaction effort, the number of design
gyrations (Ndes_g,),is selected based on traffic level. Determine Ninit, Ndesign, and Nmax for the
paving project from table 3.4.

For each initial trial asphalt content-trial aggregate gradation combination:

(1) Compact two specimens to Nmax gyrations in accordance with SHRP Method of Test
M-002 from loose paving mix aged in the short-term aging procedure (SHRP Method of
Test M-007).

(2) Determine the theoretical maximum specific gravity (Gmm) of the paving mix.

3. 7.4 Evaluate Compacted Trial Aggregate Gradations

In this step, the compaction characteristics of each trial aggregate gradation are
evaluated. Specifically, the volumetric properties of the paving mix are determined at Ndesig .
gyrations and the VMA is estimated at the four percent air voids level by shifting the
measured densification curve. Then, the density at Ninit and Nma x gyrations are evaluated to
determine the acceptability of the trial aggregate gradation as defined by the Superpave
criteria.

75
V . H ..
_and: VFAof each thai_ aggregate
...... .
:_adat_on: . .
_at:the same design air:voids... content,
:.::: :...: :.....:...-: :': ': ,., ..: . ::..:..:..: ......i ,, " i.i . ...... :.. . '.. ' .....:.:....:.:3:...i .:: .i.::.:.::.. :i: 4 ,.?: :.....:.:.._. . :..:: iii:...,:i:,:iii.i:. ..:.. :;:, :,i

3.7.4.1 Shifting Trial Aggregate Gradation Densification Curves

Trial aggregate gradations were compacted at a trial asphalt content. Evaluation of a


specific gradation requires the densification curve to be shifted as follows:

(1) Calculate the difference between air voids at Ndesi_.for the trial gradation and 4% air
voids.

Av--4 (3-15)

(2) Calculate the slope of trial gradation densification curve (see figure 3.7):

Slope= log30-loglO (3-16)


C3o-qo

(3) Calculate the horizontal shift factor:

Shift= A Va( log30 -loglO > (3 -17)


C3o-qo

(4) Horizontally shift the trial gradation densification curve. Shift each data point, Cx, with
the equation

l ogxI= l ogx + Shi f t (3 - 18)

where

Cx = measured density at x gyrations for trial gradation,

76
x = measured number of gyrations corresponding to density Cx, and

x' = estimated number of gyrations for the shifted densification curve.

Some extrapolation and tnmcation will be required depending upon the direction of the
shift. For the example shown in figure 3.8 the densification curve must be extrapolated to
Nraax. The extrapolated portion is a straight line projected using the slope of the last
segment of the measured curve. At the other end of the shifted curve several data points are
truncated such that the shifted curve begins at the same number of gyrations as the original
curve.

3.7.4.2 Determination of Estimated Design Parameters

Once the densification curve has been shifted, specific level 1 mix criteria can be
compared to properties estimated from the shifted curve. Specifically the mix is evaluated
based on density at Ninit , density at Nma x and VMA at Ndesig
..

(1) Evaluate the densification curve and determine the estimated specimen densities, Cinit ,

and Craax,at the key compaction points, Ninit, and Nmax.

(2) Compare Ci,it and Cm_, to specification requirements. Cinitmust be less than 89% and
Cm_xmust be less than 98%.

(3) Determine estimated VMA when Cdesig . equals 96%. At Ndo_ig n gyrations determine the
bulk specific gravity of specimen using Cdo_g
n and the theoretical maximum specific gravity.

Grab
= CdesignxG,nra (3 - 19 )

where Gra b =bulk specific gravity of the compacted mixture;


Cdo_gn = corrected density of the compacted specimen at Nd,_gn, and
Gram = maximum theoretical specific gravity.

77
Figure 3.7. Calculating the Slope of a Densification Curve

C30

E
E
(.9
I--
Z
LLJ
0 Clo ....
n-
LiJ
n -" I
m I
i I i m
I0 30 I0 30
I_0(3N

Figure 3.8. Shifting the Densification Curve of a Trial Asphalt Content

f
EXTRAPOLATED __ I

ESTIMATED DENSIFICATION / /
CURVE FOR GRADATION AT_ / /

Z
W
U

a.W _ _ DENSlFICATION CURVE FOR

SHIFT FACTOR
TRUNCATED
I

LOG N NMAX

78
(4) Calculate Va and VMA at Ndesig n with the following equations:

Va=10 0 G_-Gmb (3-20)


Gmm

and

VMA =i 00- GmbPs (3-21)


Gsb

(5) Estimate VMA at four percent air voids and compare to the VMA requirements (Table
3.2).

(6) Determine the difference in air voids content (AVa) of the compacted specimen from the
design level of 4 percent:

A Va:4 - Va (3-22)

where V a = air voids content of the trial blend at Ndesig n gyrations.

(7) Estimate the change in asphalt content (APb) needed to change the air voids content to 4
percent:

A Pb=-A VaXO .4=-O .4 (A Va) (3-23)

where AP b -- change in asphalt content; and


AVa = required change in air voids content from step (6) above.

(8) Estimate the change in VMA (AVMA) caused by the change in the asphalt content
(APb) determined in step (7) with the appropriate equation in table 3.6.

IIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIII II II IIIIIIIIIIIIIIIIIIIIIIIIIIIIII I IIIIIIIIIIII

(9) Calculate the VMA at the design level of 4 percent air voids by the equation

VMmdesig n = VMA trial + A VMA (3 - 2 4 )

79
where VMAdesig n = the VMA estimated at a design air voids content of four percent;
and
VMAuia_ = the VMA determined at the initial trial asphalt content.

Table 3.6. Calculation of AVMA

Va For Initial Trial Asphalt Content AVMA

> 4.0% 0.2xAV a

< 4.0% -0.1xAVa

In the last step the design aggregate s_ucture or gradation is selected from amo_ the
several trial aggregate gradations by comparing the estimated volumetric properties of the
adjusted (to four percent air voids) paving mixes to the Superpave criteria for VMA and
density.
IIIIIIIIIIIIIIIII I IIIIIIIIIIIIIIIIIIIIII I

(10) Compare estimated volumetric properties at the adjusted design asphalt content to the
Superpave criteria summarized in table 3.7 and select as the design aggregate structure the
trial aggregate gradation that best satisfies the criteria.

Table 3.7. Summary of Volumetric Design Criteria

Volumetric Property Superpave Limit

V_ at Ndesig
n 4%

VMA at Nd_i8 n Set by nominal maximum size of aggregate (Table


3.2)
Cinit < 89.0% of G.,.,

Cmax < 98.0% of Gram

Table 3.8 presents an example of the selection of a design aggregate structure from
among three trial aggregate gradations.

The top portion of table 3.8 presents measured compaction densities and volumetric
properties for specimens prepared for each trial aggregate gradation at the initial trial
asphalt content. Note that none of the specimens had an air voids content of exactly 4
percent. Therefore, the procedures described above must be applied to (1) estimate the
design asphalt content at 4 percent air voids, and (2) obtain adjusted VMA and compaction
density values at this estimated asphalt content.

The middle portion of table 3.8 presents the change in asphalt content (Pb) and VMA
that occurs when Va is adjusted to 4 percent for each trial aggregate gradation.

80
As an example, consider trial aggregate gradation #2 in table 3.8. Calculation of
AVa, AP b and AVMA yields

A Va=4 - (Va@Ndesign) --4 - (4.7) =-0.7 (3 -25)

APb=-O.4xAVa=(0.7) x(0.4) =0.3 (3-26)

A VMA=0,2xA Va= (0,2x (-0,7) ) =-0.i (3-27)

which are summarized in table 3.8.

Table 3.8. Selection Of a Design Aggregate Structure

Volumetric Property I Trial Aggregate Gradation Criteria


1 2 3

At the initial trial asphalt content

Pb(trial) 4.4 4.4 4.4 --

Ci.lt(trial) 88.1 86.5 87.1 --

Cdesig
. 95.9 95.3 94.7 --

Cmax(trial) 97.6 97.3 96.4 --

Va at Ndesig
n 4.1 4.7 5.3 4.0

VMAtria I 12.9 13.4 13.9 --

Adjustments to reach design asphalt content


(Va = 4% at Naesign)

AV_ -0.1 -0.7 - 1.3 --

APb 0.0 0.3 0.5 --

AVMA 0.0 -0.1 -0.3 --

At the estimated design asphalt content


(Va = 4% at Na_s_g
.)
Estimated 4.4 4.7 4.9 --
Pb(design)

VMAdesign 12.9 13.3 13.6 > 13.0

Ci.it(design) 88.2 87.2 88.4 <89.0

Cmax(design) 97.7 98.0 97.7 <98.0

81
The bottom portion of table 3.8 carries the analysis to the last stage, estimating the
compaction densities and volumetric properties at the design asphalt content and comparing
these estimates to the Superpave criteria. Again using trial aggregate gradation # 2 as an
example, the VMA and the densities at Ninit and Nmax are calculated as:

VSAdesign=VmAtrial+_ VMA=I3.4 + (-0. i) =13.3 (3-28)

Cinic(design) =Cinit(trial)-AVa=86.5-(-0.7)=87.2 (3-9.9)

Cmax(design) =Cmax(trial)-AVa=97.3-(-0.7) =98.0 (3-30)

as shown in table 3.8.

Comparing the VMA and densities at the estimated design asphalt content with the
criteria in the last column of table 3.8 shows that trial gradation # 1 does not have
sufficient VMA (12.9 % versus a requirement of >13.0%). Trial gradation has sufficient
VMA, but #2 exceeds the criterion for density at maximum gyrations (98.0 versus a
requirement of <98.0). Trial gradation #3 meets the requirements for density and VMA and
in this example is selected as the design aggregate structure.

Finally, the dust proportion, the ratio of the percent by weight of aggregate in the
design aggregate structure passing the 75p,m sieve to the effective asphalt binder content,
Vbe, is calculated and checked against the requirement in table 2.9.

3.7.4.3 Possible Remedies When Trial Aggregate Gradations Do Not Meet


Superpave Criteria

Many trial aggregate gradations will fail the VMA criterion. Generally, the Cinit and
Cm,x criteria will be met if the VMA criterion is satisfied. Two options exist to increase the
VMA of the trial aggregate gradation. First, the total gradation may be changed within the
control points to boost VMA by changing the proportion of the aggregate fractions.
Generally the VMA will increase as the gradation moves away from the 0.45 power
maximum density line. Second, the aggregate shape or texture characteristics can be
adjusted to boost the VMA.

If the trial aggregate gradations have been chosen to cover the entire range of the
gradation controls, then the only remaining solution is to find a new source of aggregate.

82
The aggregates that fail to meet the required criteria will not produce an adequately strong
aggregate structure and hence should not be used. One or more of the aggregate stockpiles
should be replaced with another material that produces a stronger structure. For example, a
quarry stone can replace a crushed gravel, or crushed fines can replace natural fines.

IIIIIII IIIIIIIIIIII I I IIIII III III I

3.8 Selecting the Design Asphalt Content

3.1, the flowII cha_ of Ievel 1 mix design. I


....II ...... .......

Once the design aggregate structure is selected, the design asphalt content is
determined for that structure. The design asphalt content is defined as the asphalt content
that provides 4 percent air voids at the Ndesig,gyrations.

Paving mix specimens, all containing the selected design aggregate structure, are
prepared at several asphalt contents and compacted to Nm,xgyrations. The asphalt content
that produces an air voids content (V,) equal to 4 percent at Ndesign gyrations is selected as
the design asphalt content.

Specifically, selecting of the design asphalt content includes the following steps:

Select four asphalt binder contents (section 3.8.1).


* Mix the binder and aggregate, and short-term age the loose mix.
Compact specimens for evaluation at each asphalt content (section 3.8.2).
Determine the asphalt content that produces an air voids content equal to four
percent (section 3.8.3).
Compare the volumetric properties at the design asphalt content to the Superpave
criteria and confirm their satisfactory compliance (section 3.8.4).

3.8.1 Selecting of Asphalt Contents for Evaluation

Determine the four asphalt contents at which to compact specimens with the design
aggregate gradation:

the estimated design asphalt content, Pb(design), from section 3.7.


asphalt content 0.5 percent below Pb (design);
asphalt content 0.5 percent above Pb (design); and
asphalt content 1.0 percent above Pb (design).

83
3. 8.2 Compact Specimens at Four Asphalt Contents

II II

This section corresponds with block (1-7) in figure 3.1, the flow chart for level 1 mix
design.

Duplicate specimens are mixed, aged and compacted at each asphalt content
determined in section 3.8.1. The compaction effort, the number of design gyrations (Ndosig,)
,
is selected on the basis of the anticipated traffic level.

(1) Determine Ninit, Ndesign, and Nmax for the paving project from table 3.4.

(2) Compact two specimens to Nm_ gyrations in accordance with SHRP Method of Test
M-002 from loose paving mix aged in the short-term aging procedure (SHRP
Method of Test M-007).

(3) Determine the theoretical maximum specific gravity (Gmm)of the paving mix.

3.8.3 Select the Design Asphalt Content Corresponding to an Air Voids


Content of Four Percent at Nacre,
,, II ....

This section corresponds with block (1-8) in figure 3,1, the flow ch_ for level lmix
design.
.............................................. I IIIIIIIIIIIIIIIIIIII IIIIII IIII IIIIIIIIIII illlll IIIIII

For each asphalt content, evaluate the densification curve and measure the corrected
specimen densities, Ci.it, Cdes_g.and C,,_, at the three key compaction points, Ni,_t, Ndesign, and
Nmag"

Determine Va, VMA, and VFA at Ndesign. This requires calculating of the bulk
specific gravity of the compacted paving mix at Nd_s_g
. from the density and the theoretical
maximum specific gravity according to equation 3-15:

Gmb: ( Cdesign) (Gram) (3-31)

where Gmb = bulk specific gravity of the compacted mixture;


Cdesig
n= density of the compacted specimen at Ndesign, and
Gmm -- maximum theoretical specific gravity.

84
Calculate V a, VMA, and VFA at Ndesig n from equations 3.1, 3.2, and 3.3:

Va=l 00 Gram-Grab (3-32)


Gmm

G,nbPs
VMA=I 00 (3 -33)
Gsb

VFA- VMA xl00 (3-34)


VMA

Plot Va, VMA, VFA, and Cdesig n versus asphalt content as shown in figure 3.9.

The Superpave software automatically generates all the graphs shown in figure 3.9.

By graphic interpolation, determine the design asphalt content (Pb) (to the nearest
tenth of a percent) at which Va is equal to four percent.

By graphic interpolation, verify that the paving mix meets the Superpave
requirements for VMA and VFA (tables 3.2 and 3.3) at the design asphalt content.

Figure 3.9. Plots of Percent Air Voids, Percent VMA, Percent VFA and Density
versus Percent Asphalt Content

12 17

>

5' z "

:
1_. 2 _
4.0
PERCENT
4.5 5D
ASPHALT
5.5 6.0
CONTENT
6.5 4.0 4.5
PERCENT
5.0
ASPHALT
5.5 6.0
CONTENT
6.5

2450

W
O 40
eo I--
23so
n- _ 2300
Z
20 uJ
I I I i _ a 2250 I I i I 4
4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5

PERCENT ASPHALT CONTENT PERCENT ASPHALT CONTENT 85


As an example of selecting the design asphalt content, consider the sample data presented
in table 3.9 and graphed in figure 3.9.

In this example, the estimated design asphalt content is 4.8 percent, the minimum VMA
requirement for the design aggregate structure (19.0 mm nominal maximum size) is 13
percent, and the VFA requirement is 65 to 78 percent.

Entering the plot of percent air voids versus percent asphalt content in the upper left
quadrant of figure 3.9 at 4 percent air voids, the design asphalt content is determined as
5.7 percent.

Entering the plots of percent VMA versus percent asphalt content (figure 3.9, upper right
quadrant) and percent VFA versus percent asphalt content (figure 3.9, lower left
quadrant) at 5.7 percent asphalt content, the paving mix is seen to meet the Superpave
VMA and VFA requirements at the design asphalt content.

Table 3.9. Sample Volumetric Design Data at Ndesig


n
Pb (%) Va (%) VMA (%) VFA (%) Density
(k_m3)
4.3 9.5 15.9 40.3 2320
4.8 7.0 14.7 52.4 2366
5.3 6.0 14.9 59.5 2372
5.8 3.7 13.9 73.5 2412

3.8. 4 Compare Volumetric Properties To Design Criteria at Ninit and Nmax

This section corresponds with block (1-7) in figure 3.1, the flow chart for level 1 mix
design.

In this section, the density requirements, Cinit and Cm_, at the design asphalt content are
checked against the requirements in table 3.5.

It is unlikely that an actual densification curve will be available at the design asphalt
content. Therefore, it is necessary to estimate it by shifting a measured densification curve
by an amount equal to the difference in the air voids contents (V,) at the design asphalt
content (4 percent by specification) and at the asphalt content corresponding to the
measured densification curve.

86
The same shifting procedures used in 3.7.4 are used to determine Cinit and Cma x.
From the shifted densification curve prepared at 4% air voids in section 3.7, determine Cinit
and Cma_. Confirm that Cin_tand Cm_xsatisfy the Superpave design requirements in table 3.5
at the design asphalt content.

This completes the selection of the design asphalt content, Pb-


I I

3.9 Evaluate Moisture Susceptibility of the Design Aggregate Structure at


the Design Asphalt Content

The procedure in this section corresponds with block (I_8)of fig_e 31I, the flow ch_
for level 1 mix design:
iii ..........
i...... ..... .........
ill ..................... ...........................
iiilii .............
ii.........

3.9.1 Evaluation of Moisture Susceptibility with AASHTO Method of Test T


283 (Superpave Alternative I)

(1) Compact six I specimens composed of the design aggregate structure at the design
asphalt content with the gyratory compactor in accordance with SHRP Method of
Test M-002. Referring to the shifted densification curve from section 3.7, apply
sufficient gyrations to produce an air voids content of 7 percent.

(2) Condition three specimens in accordance with AASHTO T 283. The remaining three
specimens are tested dry (i.e., without conditioning).

(3) Test the six specimens in accordance with AASHTO T 283.

(4) Calculate the tensile strength ratio (TSR) as follows:

TsR= S2xlO0 (3-35)


$I

where S_ = average tensile strength of dry subset, and


S2 = average tensile strength of conditioned subset.

Currently AASHTO T 283 recommends 100 mm diameter specimens, but permits 150 mm
diameter specimens.
87
(5) If the TSR is less than 80 percent, remedial action such as the use of antistrip agents
is required to improve the moisture susceptibility of the paving mix. The paving mix
must be then be retested to ensure compliance with the minimum 80 percent
requirement.

3.9.2 Evaluation of Moisture Susceptibility with SHRP Method of Test M-O06


(Superpave Alternative II)

(1) Compact six 2 specimens composed of the design aggregate structure at the design
asphalt content with the gyratory compactor in accordance with SHRP Method of
Test M-002. Referring to the shifted densification curve from section 3.8.4, apply
sufficient gyrations to produce an air voids content of 7 percent.

(2) Condition three specimens in accordance with SHRP Method of Test M-006. The
remaining three specimens are tested dry (i.e., without conditioning).

(3) Test the six specimens in accordance with Method of Test M-006.

(4) Calculate the resilient modulus ratio (MRR) as follows:

MeR= M_2 xl00 (3-35)

where MR1= average resilient modulus of dry subset, and


MR: = average resilient modulus of conditioned subset.

(5) If the MRR is less than 70 percent, remedial action such as the use of antistrip agents
is required to improve the moisture susceptibility of the paving mix. The paving mix
must be then be retested to ensure compliance with the minimum 70 percent
requirement.

3.10 Adjusting Mixture To Meet Properties

If the criteria in sections 3.8 or 3.9 cannot be met, adjustments must be made to the
paving mix. Possible areas of noncompliance with criteria include VMA, VFA, and
moisture susceptibility.

2 The environmental conditioning system requires 100 mm diameter, 100 mm high


specimens.

88
3.10.1 Voids in Mineral Aggregate (VMA)

VMA failures will be rare since the selection of design aggregate structure is based
on meeting the VMA criterion. However, if a change in the design aggregate skeleton is
required, there are three likely options:

Change the gradation. The gradation should be moved further from the maximum
density line. Changing gradation may not be an option if the trial aggregate
gradation analysis spans the full gradation control area.

* Reduce the minus 75_tm fraction. Reducing the dust content of the paving mix will
typically increase the VMA. If the dust content is already low, this is not a viable
option.

Change the surface texture, or shape of one or more of the aggregate fractions. This
option will require further processing of existing materials or a change in aggregate
sources.

3.10. 2 Voids Filled With Asphalt (VFA)

The lower limit of the VFA range should always be met at 4 percent air voids if the
VMA meets requirements. If the VFA upper limit is exceeded, then the VMA is
substantially above the minimum required. If so, the mixture should be redesigned to reduce
the VMA. There are three likely options:

Change the gradation. The gradation should be moved closer to maximum density
line.

Increase the minus 75p,m fraction. The dust content should be increased if room is
available within the specification control points.

Change surface texture and shape of aggregates. The aggregate should be changed to
incorporate material with better packing characteristics (e.g., fewer thin, elongated
aggregate particles).

3.10.3 Retained Tensile Strength or Resilient Modulus

Retained tensile strength or resilient modulus can be increased as follows:

by addition of chemical antistrip agents to the asphalt binder to promote adhesion in


presence of water; or

by addition of hydrated lime to the paving mix.

89
3.11 Final Result of a Superpave Level I Mix Design

To promote uniformity and ease of comparison among different agencies and


laboratories, paving mixes prepared by the Superpave mix design method are classified
according to the designators described in Appendix D.

The final products of a level 1 mix design are:

a design aggregate structure; and


a design asphalt content.

The level 1 mix design procedure can be used as a stand alone design for low-
volume roadways or as the starting point for either a level 2 or level 3 mix design as
required by anticipated traffic requirements. The level 2 mix design procedure is described
in chapter 4 of this manual, the level 3 procedure in chapter 5. A worked example of a
level 3 mix design, including a complete treatment of level 1 design is presented in
chapter 6.

90
4

Level 2 Mix Design (Intermediate Traffic Levels)

4.1 Introduction

The Superpave level 2 mix design builds upon the historical mix design practices
found in the level 1 mix design, and involves performance tests which measure fundamental
properties and predict pavement performance. These tests are described in chapter 1. The
tests required are the repeated shear at constant stress ratio, simple shear at constant height,
frequency sweep at constant height, and indirect tensile strength to predict permanent
deformation and fatigue cracking; and the indirect tensile creep and indirect tensile strength
to predict low-temperature cracking. These predictions are based on validated material
properties which have been linked to pavement performance.

For permanent deformation, the simple shear at constant height, repeated shear at
constant stress ratio, and frequency sweep at constant height are conducted at the effective
temperature for permanent deformation. This effective temperature is calculated from the
pavement temperatures for the project site. For fatigue cracking, the frequency sweep at
constant height, the simple shear at constant height, and the indirect tensile strength tests
are conducted at the effective temperature for fatigue cracking. This effective temperature is
also calculated from the pavement temperatures for the project site.

Predictions for low-temperature cracking are based on indirect tensile creep tests
performed at 0C, -10C, and -20C and on indirect tensile strength tests performed at
-10C. Bending beam test data from the asphalt binder test are used in conjunction with
the creep data to predict low-temperature response. A time series of historical pavement
temperatures is used to predict tensile stress and fracture.

Since the analyses are independent of each other, the test sequence for any of the
three distress types can be eliminated. Thus the design can be focussed on distress types of
primary concern. Additionally, the three shear tests can be conducted using the SHRP
shear test device, or in an intermediate shear test device designed to operate at elevated
temperatures without confining pressure.

Level 2 mixture design has been kept uncomplicated enough for routine use. Figure
4.1 illustrates the flow chart for the level 2 mix design process. In situations where level 1

91
mixture design with no mechanical property measurement is not considered to be
sufficiently reliable, level 2 mixture design should be used. Level 2 mixture design is
expected to be the most common design type used in typical highway applications. The
level 2 mix design should be used for project traffic having greater than 10 6 but less than
10 7 design equivalent single axle loads (ESALs). Level 2 mixture design will provide a
mix design that will perform as required without excessive design effort and expense.

4.2 Outline of Level 2 Mix Design Method

The procedure for the Superpave level 2 (intermediate traffic) mix design method
starts with the sample preparation and volumetric design method (level 1 design) described
in Chapter 3. The level 2 design method is applied when the design ESALs over the
anticipated service life of the pavement are projected to fall between 10 6 and 10 7.

In the Superpave mix design method, the steps needed to perform the level 2
(intermediate traffic) mix design include:

Conduct volumetric (level 1) mix design.


Select three asphalt contents bracketing the level 1 design asphalt content.
Specify distress types for evaluation.
Obtain pavement temperature history.
Calculate effective temperatures.
Run performance-based tests.
Evaluate test data to calculate material properties.
Obtain predicted traffic levels.
Obtain structural layer information.
Predict pavement performance.

Table 4.1 shows the tests required for the various distress factors. Table 4.2 illustrates the
number of compacted specimens by distress factor and test. Appendix B provides for
reference the system of specimen coding used by the Superpave software.

92
X_, _ L.

_# C roco (3J-l_ _-- u v rO __ c- qj r._E

__. f,.. c'- --_ , e7


Z r_ _ C"3 c- no ro_ _ I e.-- 0_,_ (_
O. -Ii_ C _/) Z5 --

e_l _ r_c-rO c U(IJ OPO c-ZS=

_ '_ I
_ 0 _
CO II _ qJ _ _ i =5
E cx,I
_ I

_._ , ._,..__C_ ._ Od I
a_ "_'<_ -_,_'_"x'_ "_ _'
_ _ _ _: co _ c- o
o x _ _ ,
..1,.
_-_:
0 _j_r__ . ":'_ _ ___ '-
_

u') _") c- ,,_ (1) -- .,:_ _.l. _ L0 c- q._ __ I

,..i =

_,_ . .__ _ _ _ _ ...-.


<._. _ _-_
_ , -I-_r_ _ _.)c- r_ _0 e- _,0__ CO _ c- O_F =

,IV I ,,_J_ - I_) _ I_ (_ I_ I_ r

_I_ . _ _ c- or--.- _,,_ _ _ N c- _ __ _._ _ _ r_ O..c- c-


"-s _.,_ c" ._ U3 _,3 O _ _.,0 Of,-- .._ (3J _,9 (]._ (3._ 4--
_ _ _ -- _ _ _ " _ _ _ _ _ _ _ ' _ O

_=_ u._ ,'--_ u u <_ ....


_,_ rO c- _ ..C= ""s "s _ _.. _.. c- "s rO c-
__ o_ _'_ _- -_._ _
.,..i _ _ _ _ r'_._ _ U
_,_ c- _ _ '_',L._"C_ _ _= _/_ _.) _ 4-- tO O'_ C"q-- C.Jl--- _ C_4-- rO

r_

93
Table 4.1. Performance Tests For Level 2 Mix Design

Permanent Deformation Fatigue Cracking Tests Low-Temperature


Tests Cracking Tests

Repeated shear at constant stress Simple shear at constant Indirect tensile creep
ratio (tertiary creep)* height at Telf at 0C, -10C and -20C.

Simple shear at constant height at Frequency sweep at Tar Indirect tensile strength
Teff at - 10C.

Frequency sweep at T_ff Indirect tensile strength Creep stiffness (S) and
at Tefr slope (m) of binder from
bending beam test

*Conducted to insure that excessive permanent deformation will not occur early in the pavement service life.

Table 4.2. Compacted Specimen Requirement for Level 2 Mix Design


(Per Asphalt Content)

Method of Test Testing Temperature (C)

-20 -I0 0 Tea(PD) T_(FC)

Repeated shear at constant stress ratio 2 (at To)


(tertiary creep)

Simple shear at constant height 2* 2*

Frequency sweep at constant height 2* 2*

Indirect tensile strength 2


(50 mm/min loading rate)

Indirect tensile creep 3* 3* 3*

Indirect tensile strength 3*


(12.5 mm/min loading rate)

*Tests performed on same specimens.

94
4. 2.1 Historical Air Temperature

Superpave uses a 10-year block of historical temperatures in performance prediction.


Unlike the statistical air temperature information used for asphalt binder grade selection in
level 1 design, a database containing 10 years of historical daily minimum and maximum
temperatures for all of the United Stated is impractical. A library of historical air
temperature for each state is available with the Superpave software. Mix designers can
install temperature files for areas of interest, thus reducing the volume of unneeded
information.

4.2.2 Historical Pavement Temperature

Historical air temperatures are converted into historical pavement temperatures using
the Federal Highway Administration Environmental Effects Model which is embedded in
the Superpave software. The output of the Environmental Effects Model used by the
Superpave software is limited to pavement temperatures. Temperatures of other layers and
other model outputs are not used.

4.2.3 Pavement Temperature Converted Into Effective Temperature

The level 2 mix design predicts performance for permanent deformation and fatigue
cracking by collapsing the entire pavement temperature history into a single effective
temperature at which the material properties are measured and for which pavement
performance is predicted. The effective temperature differs for permanent deformation and
fatigue cracking. The Superpave software calculates both effective temperatures which are
used as test temperatures for the performance-based tests.

4.2.4 Calculate Material Properties

The Superpave software accesses data files created by the data collection software
which is associated with the Superpave performance test equipment. Testing machines have
stand-alone software to control execution of the performance-based test protocols. The
Superpave software processes the data files using embedded software to obtain material
properties.

4.2.5 Predict Pavement Performance

The Superpave software predicts pavement performance using material properties,


traffic, pavement temperature, and pavement structure, and displays results in terms of
distress versus time.

95
4.3 Equipment

The equipment required to prepare and test compacted specimens is as follows:

Pan: metal, flat bottom, for heating aggregates.

Pans: metal, round, approximately 4-1 capacity, for mixing asphalt and aggregates.

Oven and hot plate: electric, for heating aggregates, asphalt, and equipment as
required. Oven must also be capable of maintaining the temperature for short-term
aging (SHRP Method M-007).

Scoop: for batching aggregates.

Containers:, gill-type tins, beakers, pouring pots, or sauce pans, for heating asphalt.

Thermometers: armored, glass, or dial-type with metal stem, IOC to 230C, for
determining temperature of aggregates, asphalt, and asphalt mixtures.

Balance: 5-kg capacity, sensitive to 1 g for weighing aggregates and asphalt.


Balance 2-kg capacity, sensitive to 0.1 g for weighing compacted specimens.

Mixing spoons: large, or trowel: small.

Spatula: large.

Mechanical mixer: (optional), commercial bread dough mixer 4-1 capacity or larger,
equipped with two metal mixing bowls and two wire stirrers.

Gloves: welders, for handling hot equipment. Gloves: rubber, for removing
specimens from water bath.

Marking crayons: for identifying test specimens.

SHRP gyratory compactor: capable of providing a consolidation pressure of 0.60


MPa, an angle of gyration of 1.25 degrees, and speed of gyration of 30.0 rpm, as
described in SHRP Method of Test M-002.

Cylindrical molds: large enough to accommodate the following specimen size


requirements: 150 mm diameter, 150 mm maximum height and 90 mm minimum
height.

Extrusion jack or arbor press: for extruding compacted specimens from mold.

96
Superpave shear test device: meeting the requirements in SHRP Method of Test M-
003, or capable of conducting the shear tests required.

Indirect tensile test device: meeting the requirements in SHRP Method of Test M-
005, or capable of conducting the indirect tensile tests required.

4.4 Selecting of Mixture Materials for Volumetric Design


::::::::::::::::::::::::::::::::::
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
::::::::::::::::::::::::::::::::
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::;
:::::::::::::::::::::::::::::::::::::

iiiiiiili_iiiiii_iiiiiiiiiii_
i!ii___: _iiiiiii_iiiiii_iiiiiiiiiii_ii_i!i_i_iiiiii_iiiiiiiiiiiiiiiiiiiiii_iiii
]]]]]]]]]-rn]]-l]]]
i!iiiiiii!il
ii!i!iiiiiii
iiiiii
iiiiiiiiiiiiiii!iiiiiiiiiiiiiiiii
........
iliiiiiiiiiiiiiii
iiiiii!iiiiiiiiiiiiiiiiiiiiiiiiiiiii!i!i
iii!iiii!i!iiiiiii_iiiiiiiiiii
!iiiiiiiiiiiiiiiii
M]]] ........................................................................................................................................

The first important step in the level 2 mix design procedure is selecting materials
suited to the demands of traffic and environment expected over time for the paving project.

(1) Select the fine and coarse aggregate for trial blends, in accordance with the details
provided in chapter 2 of this manual.

(2) Select the performance grade of the asphalt binder in accordance with the details
provided in chapter 2.

(3) Conduct the (optional) net adsorption test to evaluate asphalt-aggregate


compatibility, in accordance with the details provided in section 2.2.3.

(4) Conduct the trial asphalt content and trial gradation evaluations and determine the
design aggregate structure and the design asphalt content based on volumetric
properties requirements, in accordance with the procedures provided in chapter 3 of
this manual.

Note: The design traffic level (Nd_ign)for the level 2 mix design is between 10 6 and 10 7
ESALs as discussed in chapter 3. N_,itand N,,_ are presented in table 3.4.

4.5 Selecting of Asphalt Contents to Yield 3, 4, and 6 percent Air Voids

_:_:_:''":i:i_:i:i:i:i:!:!:i
:i:i:i
:i:i:i:i:i:
_:i:i:i:i:i":i:i:i:i:!:!:i:i:i:i:i:i:!:i:i:i:i:i:i:i:i:i:i:!:!:i:i::'i:i:i:i:
i"i:i:i:i
_:i:i:i:i:i:!:::":!:i_:i:i:i:i:i:i
:i:i:i:i:!:i:i::'i_:i:!:!:_:i:i:i:i:i:i:i:i:i:!:i:_:!:i:i:i:i:i:i:i:
i:!:i:i:
i:_:i:i:i:i:i:'
":i:i:i:_:i:i:i:i:i:i_:i':
":ii i ?"i:i"i:_:i:i:i:i:"':i:i
i i::":':i i:""i ::_:_:i
i:i:i::""
i i"'"i!:_:i:i:i:i:i:i:i
!:i:i:__:i"iii :'"i __!:!:::":i:!:i:i
_i i :iii i iiii
iii_iiii__ i_i__iii_ !i_ii _ii __ ii_ i_ i_ii _ _ i_!__::::::
iiiiii_Niiii_N
ii_iiiii_Niiii_iliN_i!ii!i_iNiiiiiiiiiiiiiiiill
@]....................
iiiiiSiiiiiii
iiii!iii!iiiii
iiiiiiiiiiiiiiiiiiii
!!iiiiiiiiiiiiiiiiiiiiiiii!
]]'IT''TTT]]]]])]I]I)))]]]]r
iiiiii!ii!jiiiiiiiiiiiiiiiiiii
iiiiiiiiiiiiiii
iiiiiiiiiiiiiiii
iiiiiiiiiii
iiiiiiiiiiiiiiiiiii
iiiiiiiiiiiiiiiiiiiiiiiiiiiii
iiii
iiiiiiiiiiiiiiiiiiiiiiiiii
iiiiiiiiiiiiii
(lrlT]ll]]]T']..................
iiiiiiiiiiiiiiiii
iiiiiiiiiiiiiiiiiiiiii!iii
iiiiiiii
iiiiiiiiiiiiiiiiiiii
(I......................
1...........................................
""...................
".......................
""".....................
"...................................................................................................
"................
".....................

(1) Refer to section 3.8.3 for the level 1 volumetric mix design. At the design number
of gyrations (Nd,_g,) volumetric properties are calculated for each asphalt content as
shown in table 3.9. Air voids, VMA, and VFA are plotted for each asphalt content
as shown in figure 3.9.

97
(2) From the plot of percent air voids versus percent asphalt content (figure 3.9),
determine the asphalt contents at three, four, and six percent air voids.

(3) Designate the asphalt contents at three, four, and six percent air voids as high,
design, and low, respectively.

4.6 Determining the Effective Temperature for Fatigue Testing


Several approaches were investigated to establish an effective temperature for fatigue
cracking (Teff(FC)) analogous to that illustrated in section 4.7 for permanent deformation. In
the level 2 mix design procedure Teft(FC) is defined as a single test temperature at which an
amount of fatigue damage would occur equivalent to that measured by considering each
season separately throughout the year.

Because the fatigue cracking mechanism consists of two separate phases, crack
initiation and crack propagation, procedures for selecting an equivalent temperature were
complex and inappropriate for use in routine mix design applications. A simplified
approach was used in which the total fatigue damage was computed with the Superpave
fatigue cracking model on a month-by-month basis for 12 Long-Term Pavement
Performance GPS pavement sections with varying levels of fatigue cracking. The average
monthly total fatigue damage was then calculated and used to determine a single test
temperature for each pavement that would result in an equivalent amount of fatigue
cracking.

For simplicity, the fatigue damage equivalent temperatures were compared to the
mean annual pavement temperatures for each pavement. This analysis showed that a good
value for the equivalent fatigue damage temperature, T_ff(FC), can be estimated from the
mean annual pavement temperature with the following equation:

T,_FCO=O.8(MAPT) -2.7 (4-1)

where T_FC) = the effective test temperature in C for fatigue cracking; and

MAPT = the mean annual pavement temperature in C calculated at one-third of


the depth of the pavement layer from climatic data for its geographic
location.

iNNi iiiiii iii ii! Niiill iiiiNi Nii !i i #ii iiii i i iiiNN i i iii i iiN
_ _ _ii_!ii!!ii!_iiiiii!!i!ii_iiiiiii_.!!iiiiiiiiiii!!!!i!i_iiii!!!!i_iiiiiiiii!ii_iiiiiiiiiiiiiiiiiiiiiii!!!!!!_iiiiiii!!!!_ii
..............................................................
i............................................................................................................................................................................
i..................................................................................
i:i
...........................................

T_FC) is automatically calculated for level 2 mix designs through menu-driven


routines contained in the Superpave software.

98
4.7 Determining the Effective Temperature for Permanent Deformation
Testing Including Tertiary Creep

The effective temperature for permanent deformation, Te_(PD), is defined as a single


test temperature at which an amount of permanent deformation would occur equivalent to
that measured by considering each season separately throughout the year. Te_PD) is the
temperature at which the tests to evaluate permanent deformation are conducted. This
approach simplifies testing, but is less accurate than the environmental effects model used
with the level 3 mix design approach.

T,n(PD) is automatically calculated for level 2 mix designs through menu-driven


routines contained in the Superpave software. The following sections illustrate the
methodology of the calculation of T,ft(PD) used in the software.

For the desired critical depth, Zcr , the T,gPD) for the level 2 mix design is
calculated from the following equations:

I",#(PD ) =30.8 -0.12Z_r +0.92MAA Taem,, (4-2 )

where

T,fr(PD) = effective temperature (C);

Zcr = critical depth within the mix layer in question (mm); and

and

MAATdesign=MAATaverage+K OMAA T (4-3)

where

MAAToverag_)= average mean annual air temperature computed from historical data;

O'MAAT = standard deviation of the distribution of mean annual air temperature


for the geographic location; and

K_ = value computed from normal probability tables and related to the


designer's selection of an appropriate reliability level (R) desired for
the project.

Appropriate values of K_ for given reliability levels (R) are given in table 4.3:

99
Table 4.3. Values of K_,

R (Reliability) K_

50 % 0.000

75 % 0.674

85 % 1.037

90 % 1.282

95 % 1.645
, , .

99 % 2.327

As an example, consider a surface course mix to be placed on a major interstate


highway near Richmond, Virginia. For these conditions, the following values have
been selected:

Zcr = 25 mm (surface course analysis),

MAAT = 14.4C,

R = 95% (high reliability desired for interstate-type pavement), and

c MAAT= 3-0C (computed from historic temperature data).

Therefore,

MAA T_e_ =MAA T average+r . OMAAT (4-4)

MAATaeae.n = 14.4 +(1.645)(3.0) (4-5)

MAA T_,_s,, =19.3 C (4-6)

The design Tefr(PD) value is therefore

Te_(PD ) =30.8 -(0.12)(25) +(0.92)(19.3) (4-7)

T_PD) =45.6 C (4-8)

100
4.8 Tertiary Creep Evaluation

The tertiary creep evaluation is a screening process to identify a mix that exhibits
tertiary plastic flow leading to gross mix instability. If the mix fails the screening test, it
will be necessary to either make adjustments to the mix proportioning or to redesign the
mix completely. If the mix passes the screening test requirements, the mix may be subjected
to the battery of tests for the characterization of permanent deformation, fatigue cracking,
and low-temperature cracking.

Figure 4.2 illustrates conceptually the relationship between tertiary plastic flow and
the performance prediction model established in the Superpave software. For a particular
test temperature, the laboratory relationship between the log of the permanent strain, ep, and
the log of the number of load repetitions, N, for a specific mix may actually result in the
relationship shown in figure 4.2. For this case, it is obvious that tertiary plastic flow leading
to gross mix instability is present. However, the Superpave performance model analysis
uses a linear model, or steady-state secondary creep, (i.e. the slope S) shown in figure 4.2.
Therefore, analysis of permanent deformation using the test methods in table 4.2 and the
Superpave performance model are not sufficient to identify a mix design that will exhibit
tertiary creep under certain circumstances.
I

The procedure outlined in this section e_esponds with _iocks (2_3)and (2i4)_ figure
4,1; the flow chart for the Superpave leveI 2 mix designi Steps (2) thro_h (10) are
graded by the Supe_ave so,ware. ::
I................. I I .......... I I............ I I Ill Illll : iliiillll II I I I

Primary Procedure

(1) Prepare two specimens at the high asphalt content as determined in section 4.5 and
compact to 3 percent air voids at Ndesig. gyrations, using SHRP Method of Test M-
002.

(2) For the mix type and layer thickness in question, select an appropriate value of Zcr.
(This can be simplified by selecting Zcr= 25 mm for a surface course layers; Z_r=
62.5 mm for a binder course; Z_r= 100 mm for a base course; etc.).

(3) Determine the mean annual air temperature (MAAT) and the corresponding standard
deviation (_MAm')for the design site location.

(4) Determine the design traffic repetition value N(des), and select appropriate values for
the design reliability (R) and the K_.

(5) Compute the Telr value for the pavement location.

101
(6) Plot the point (N(des)-Teg) on Figure 4.3 and determine the appropriate Curve ID
letter. If this point plots "beyond Curve M", go to the alternative procedure, steps
(8) through (10).

(7) Determine the recommended control temperature, To, at which to conduct the
repeated shear test at constant stress ratio (SHRP Standard Practice P-005) from the
following table:

Curve ID Tc(C) Curve ID Tc(C)

A 29 H 49

B 32 I 52

C 35 J 54

D 38 K 57

E 41 L 60

F 43 M 63

G 46

Figure 4.2. Typical Relationship of Plastic (Permanent) Strain, ep, and Load
Repetitions, N, For Mixes with Tertiary Creep (Plastic Flow)

TEMP: CONSTANT
TERTIARY (PLASTIC)
FLOW

z
n-
l---
or)
O

O')
..J


0
d
R LINEAR
MODEL

T
LOG N (REPETITIONS)

102
8"

_ _ o
0 _

0 t" "->

"_ -_ co I

r_ ,,, co

_ rn _

.._

o g o
o o
o oo o o o o
o o o o o
o
0
0
o o
suo!_!_edeEI uS!seG -- (sep) N

103
At this To, the repeated shear test is conducted for 20,000 repetitions, at the shear
and compressive stresses selected from table 4.4; this requires 4 hours of testing.

Table 4.4. Maximum Shear and Compressive Stress Levels

Base Condition Maximum Shear (x) And Compressive


(_v) Stress Levels (kPa) at Asphalt Binder Content

Above Design Design Below Design

T O"v "_ O"v 1: O"v

Weak 84 119 63 98 49 56

Strong 98 175 84 105 56 91

Note: A weak base is defined as an unbound granular or crushed stone material (i.e; new construction); a strong base is defined as
an existing asphalt concrete or portland cement concrete pavement, a cement-stabilized or asphalt-stabilized base, or a strong
crushed stone base material (i.e., a resilient modulus of 560,000 kPa or greater).

Alternative Procedure

This procedure is employed when a value of Tc greater than 63C would be required for the
primary procedure.

(8) Plot the point (N(des)-Teff) on figure 4.4 and determine the appropriate Curve ID
letter. If this point plots "beyond Curve E," see the instructions following step 10.

(9) Determine the required number of repeated shear test repetitions, Neq, and the
expected test period from the following table:

Curve ID N_q Time (hr)

A 5,000 I

B 10,000 2

C 20,000 4

D 40,000 8

E 120,000 24

(10) Conduct the repeated shear test at constant stress ratio at a test temperature of Tc =
63C and the stress and compressive stresses determined from table 4.4 for the
number of test repetitions determined in step (9).

104
105
Points plotting beyond Curve ID "E" on figure 4.4 will generally be very difficult to test
due to the combinations of high temperature, stress state, and the required repetition level
(i.e., time of test). For these rare cases, it is suggested that a reduced design period be
selected and the process repeated until a solution is feasible (step (7) or step (9)). When
this case occurs, the mix design engineer should realize that full assurance of the mix
quality cannot be made for the original design period desired.

Data from the repeated shear at constant stress ratio test is plotted as shown in
Figure 4.2. Suitability of the mixture is determined qualitatively; if the mixture exceeds the
linear phase and enters the tertiary flow region the mixture is judged unsuitable for the
intended purpose.

Note that tertiary creep evaluation is performed only on the uppermost pavement
layer at the highest asphalt content to be evaluated.

4.9 Permanent Deformation and Load-Associated Fatigue Property


Characterization

i_ii?::_
_!_!ii_ii_i_iiiiiiiii_iiiii_ii!iiiiiiiii!_:::!!ii_i_iiii_iii_iiiiiiiiiiiiiii!i!i_!i_
_ii_i_iiiiiiiiiiiiiii::ii:::!_i_i_i_i_iiiiiiiiii::ii:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
.............
::::::::::::::::::::::::::::::::::::::
...................

i iiii iiii iii iiiiii !iii iiii i iiii !i iiii@ iiii iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii !!
IIIIIIIIUIIIIII HHUIIIIIIIIIIIIIIIIIIIIIIIIII

The procedures presented in this section determine by appropriate test methods the
material properties needed to predict the mix's potential for permanent deformation and
load-associated fatigue over its service life.

(1) Compact (by SHRP Method of Test M-002) three sets of four specimens each to 7
percent air voids at the high, design, and low asphalt contents determined in section
4.5. Twelve specimens are prepared (4 specimens per asphalt content times 3 asphalt
contents = 12).

(2) Six specimens (2 specimens per asphalt content times 3 asphalt contents = 6) are
selected for testing for permanent deformation and load-associated fatigue properties.
The other six specimens will be used for determining the indirect tensile strength in
step 6.

(3) For each asphalt content, test one pair of specimens by performing the frequency
sweep at constant height, in accordance with section 7.5 of SHRP Practice P-005.
Nine frequency sweeps are conducted at 10, 5, 2, 1, 0.5, 0.2, 0.1, 0.05, and 0.02 Hz.
Each series of sweeps is conducted at the T,n(FC) determined for fatigue cracking in
section 4.6. Record the axial deformation, shear deformation, axial load, and shear
load during the load cycles at the rate specified in SHRP P-005 and in a format
suitable for analysis by the Superpave software.

106
(4) Using the same pair of specimens tested in step 3 conduct the frequency
sweep at constant height with section 7.5 of SHRP Practice P-005. Conduct
the same nine frequency sweeps as specified in step 3. Each series of sweeps
is conducted at the Terf(PD) determined for permanent deformation in section
4.7. Record the same measurements specified in step 3.

(5) Using the same pair of specimens tested in steps 3 and 4, conduct the simple
shear at constant height test in accordance with section 7.4 of SHRP Practice
P-005 at Teff (PD) and Teff (FC), respectively. At both test temperatures, record
the axial deformation, the shear deformation, the axial load, and the shear
load at a rate of about 10 data points per second. The data recorded must be
in a format suitable for analysis by Superpave software.

(6) For each asphalt content, test the remaining pair of specimens prepared in
step 1 by performing the indirect tensile strength test in accordance with
SHRP Method of Test M-005 at Teff (FC). Record the maximum peak load at
a ram load rate of 50mm per minute in a format suitable for analysis by the
Superpave software.

4.10 Predicting Permanent Deformation and Load-Associated Fatigue


Cracking at the Pavement Design Life

The procedure outlined in this section corresponds with block (2-10) in figure 4.1, the
flowchart for level 2 mix design.

(1) The data recorded in section 4.9, steps 3 through 6, are transferred to the
Superpave software in a format suitable for analysis.

(2) The Superpave software predicts permanent deformation for a standard load
applied to a pavement structure at the effective temperature. Stresses are
calculated within each pavement layer using pavement temperature and
mixture stiffness obtained from the frequency sweep test data. Permanent
deformation of the pavement layer is calculated using simple shear test
results, and total permanent deformation (rut depth) at the pavement surface
is calculated versus time.

(3) Fatigue cracking is estimated using an effective temperature for a single


season, as calculated for fatigue cracking in section 4.6. Stresses within each
pavement layer are calculated for the layer stiffness representative of the
temperature under a standard load. Dissipated energy is calculated for each
single season, and the cumulative dissipated energy is used to estimate the
percent of fatigue cracking over the anticipated or designated service life of
the pavement is predicted.

107
4.11 Low-Temperature Cracking (Non-Load Associated) Property
Characterization

The procedures in this section correspond with blocks (2-5 and 2-9) in figure 4.1, the
flow chart for level 2 mix design.
.......... IIIIIIIII I IIIIIII I III IIIIIIII IIII II II I IIIII IIIIIIIIIIIIII IIIIIIIIII III I IIII iiiii iiiiiii ii llllllll

The procedures outlined in this section determine the material properties by


appropriate test methods in order to predict the mix design's potential to experience low-
temperature cracking over its service life.

(1) Compact (by SHRP Method of test M-002) three sets of three specimens each
to seven percent air voids at the high, design, and low asphalt contents as
determined in section 4.5. Nine specimens are prepared (3 specimens per
asphalt content times 3 asphalt contents = 9).

(2) For each asphalt content, test three specimens by performing the indirect
tensile creep test at -20C, - 10C and 0C and the indirect tensile strength
test at -10C in accordance with section 9.7 of SHRP Method of Test M-005.
For each specimen, continuously record the load and the vertical and
horizontal deformation measurements in a format suitable for analysis by the
Superpave software.

4.12 Predicting Low-Temperature Cracking at the Pavement Design Life

The procedure outlined in this section corresponds with block (2-11) in figure 4.1,
the flow chart for Level 2 mix design,
II II I I I I I I I

(1) The data recorded in section 4.11 are entered into the Superpave software in
a format suitable for analysis.

(2) Low-temperature cracking is predicted using low-temperature creep behavior


and low temperature fracture strengths. Predictive models are used with
statistical weather data to predict stresses caused by overnight temperature
cycles. Fracture mechanics is used in conjunction with tensile strength and
predicted tensile stress to determine the advancement of cracking. Estimates
are obtained of crack spacing with time.

108
4.13 Selecting the Optimum Asphalt Content

The predicted values of permanent deformation (rut depth), fatigue cracking


(percentage of pavement area) and low-temperature cracking (crack spacing) are used to
select the optimum asphalt content.

The magnitudes of the predicted values for the distress factors determined in
sections 4.10 and 4.12 are plotted against the asphalt contents tested in sections 4.9 and
4.11. Figure 4.5, illustrates the plots. Table 4.5 presents suggested acceptable levels of
performance for each distress. These may be adjusted as the design agency acquires more
data or totally replaced by the design agency's specific levels. The design agency's
acceptable levels for permanent deformation, fatigue cracking, and low-temperature
cracking are identified on the vertical axis of the graphs and projected to the right to
intersect with the plot. A vertical line is projected at this point of intersection perpendicular
to the horizontal axis as shown in figure 4.5.

Table 4.5. Suggested Acceptable Levels of Performance


Level 2 Mix Design

Distress Factor Acceptable Level

Permanent Deformation
Reliability (%) 95
Average rut depth (in) 10

Fatigue Cracking
Reliability (%) 90
% of Area 10

Low-Temperature Cracking
Reliability (%) 80
Crack Spacing, m >150

109
Figure 4.5. Selection of Optimum Asphalt Content Range Satisfying Three Distress
Factor Conditions

15.0 -

12.5

(ram) 10.0 Accepta

7.5

Rut Depth
5._ 6.0 6.5
Low Design High
Asphalt Content
(a)
2O

15

FatigueCracking
(PercentArea) 10 Acceptable_._ "

0 i
5.5 6.0 6.5
Low Design High
Asphalt Content
(b)

OI-
Cracking
Low-Temperature . Accepta bJe_ _

(Crack spacing, m) 150 ,-_]_,_.

300
5.5 6.0 6.5
Low Design High
Asphalt Content
(c)
110
Asphalt contents to the left of the vertical line (figure 4.5, top) as shown by the
shaded area satisfy the permanent deformation requirement. Asphalt contents to the right of
the vertical line (figure 4.5, middle and bottom) as shown by the shaded area satisfy the
fatigue and low temperature requirements.

The dashed area indicates the range of asphalt contents that will satisfy all three
distress factor requirements related to the design agency's acceptable levels. Select an
asphalt content within this band and designate as the optimum asphalt content to be used
for mix production.

If only two distress factors are of prime importance to the designer, a similar
procedure is used (figure 4.6). For this example, permanent deformation and fatigue
cracking are being considered. The dashed area indicates the range of asphalt contents that
will satisfy the design agency's acceptable levels for these two distress types. The optimum
asphalt content is selected within this range.

If one distress factor is of primary concern the following approach is used. Refer to
figure 4.5 or 4.6. Using permanent deformation as an example, the asphalt contents to the
left of the vertical line (as shown by the shaded area) satisfy the permanent deformation
requirement. The same approach is used for fatigue or low temperature cracking, except
the shaded area to the right of the vertical line in figure 4.6 satisfies the requirement. The
optimum asphalt content may be selected within this range identified by the shaded area.

If a range of acceptable asphalts cannot be established to select an optimum asphalt


content that satisfies all the distress factors evaluated, the mix proportioning can be
adjusted or a modifier or modified binder should be considered in the mix design.

The Superpave software provides the comparative plots discussed in this section.
They are generated from the test input data and the design agency's acceptable level of
performance.

4.14 Treatment of Modified Materials and Mixtures

If a modifier is used to enhance the mix performance capability, the procedures


outline previously for permanent deformation and low-temperature cracking apply also to
modified mixtures. The prediction models used in Superpave will apply to both
conventional and modified asphalts and mixtures for evaluating pavement performance in
terms of permanent deformation and low-temperature cracking.

However, if fatigue is the principal distress type addressed by a modifier, flexural


beam testing is recommended in place of the frequency sweep, simple shear at constant
height, and indirect tensile strength test procedures. No prediction capability is available
for this situation with the Superpave models. A comparison of the cycles to failure can be
made between the conventional and modified mixture.

111
Figure 4.6. Selection of Optimum Asphalt Content Range Satisfying
Two Distress Factor Conditions

15.0 -

12.5 -
Rut Depth
(mm) 10.0 Acceptable

::::::::::::::::::::::::::::
::::::::::::::::::::::::::: ....
::::::::::::::::::::::::::

5.5 6.0 6.5


Low Design High
Asphalt Content
(a)
20

15
Fatigue
(PercentCracking
Area) 10 Acceptable_._

......

0 - _ __::::i!
5.5 6.0 6.5
Low Design High
Asphalt Content
(b)

112
5

Level 3 Mix Design (High Traffic Levels)

5.1 Introduction

The Superpave level 3 mix design is similar to the level 2 mix design procedure,
except that a more complete set of performance-based mixture properties are obtained, and
the performance models used to predict fatigue and permanent deformation are more
comprehensive in level 3 than in level 2.

To complete the characterization of material properties, two tests are used in


addition to the performance-based tests from level 2. Figure 5.1 shows the level 3 mix
design flow chart. The following tests are performed:

repeated shear test at constant stress ratio (tertiary creep);


frequency sweep at constant height;
simple shear at constant height;
uniaxial strain;
hydrostatic state of stress (volumetric);
indirect tensile creep; and
indirect tensile strength.

The simple shear at constant height, repeated shear at constant stress ratio, frequency
sweep at constant height, indirect tensile creep, and indirect tensile strength tests are
performed in the same manner as in level 2 mixture design except they are conducted over
a range of temperatures and must utilize the SHRP shear test device or its equivalent.

In the level 3 design, the hydrostatic state of stress (volumetric) test and the uniaxial
strain test are used to measure the nonlinear elastic behavior of the aggregate skeleton.
Nonlinear elastic behavior comes from aggregate particles. As the particles move in contact
with each other, the stiffness of the aggregate skeleton increases. Mixture performance
predictions for the level 3 mix design are performed in a similar manner as in level 2. The
main differences from level 2 are the following:

Material property characterization is more complete including nonlinear elastic


behavior for level 3 mix design.

113
Weather data and lower layer moduli are evaluated in a more rigorous manner for
level 3 mix design. The year is divided into seasons, each with specific pavement
temperatures and lower layer moduli.

In terms of other input data, predicting pavement performance is done in the same manner
as in the level 2 mix design discussed in chapter 4.

5.2 Outline of Level 3 Mix Design Method

The procedure for the SHRP level 3 mix design method starts with the sample
preparation and volumetric design method (level 1 design) described in Chapter 3.
However, the design equivalent single axle loads (ESALs) related to the gyratory compactor
described in chapter 3 are usually greater than 10 7, although this value can be increased or
decreased at the discretion of the agency.

With the Superpave mix design system, the steps to performing the level 3 mix
design include the following:

Conduct volumetric mix design;


Select three asphalt contents bracketing the level 1 design content;
Obtain historical pavement temperatures;
Run performance-based tests;
Evaluate test data to obtain material properties;
Obtain predicted traffic volume;
Obtain structural layer information; and
Predict pavement performance.

Tables 5.1 and 5.2 identify the tests and the number of compacted specimens
necessary to perform a level 3 mix design. Appendix B provides for reference the system of
specimen coding used by the Superpave software.

5.2.1 Historical Air Temperature

The Superpave software uses a 10-year block of historical temperatures in


performance prediction. Unlike the statistical air temperature information used for asphalt
binder grade selection in level 1 design, a database containing 10 years of historical daily
minimum and maximum temperatures for all of the United States is impractical. A library
of historical air temperatures for each state is provided with the Superpave software. Mix
designers can install temperature files only for areas of interest, thus reducing the volume of
unneeded information.

115
5. 2. 2 Historical Pavement Temperature

Historical air temperatures are converted into historical pavement temperatures using
the Federal Highway Administration Environmental Effects Model which is embedded in
the Superpave software. Model output used by Superpave is limited to pavement
temperatures. Temperatures of other layers and other model outputs are not used.

5.2.3 Calculate Material Properties

The Superpave software accesses data files created by the data collection software
which is associated with the performance tests. Testing machines have stand-alone software
to control execution of the performance-based test protocols. The Superpave software
processes the data files using embedded software to obtain material properties.

5. 2. 4 Predict Pavement Performance

The Superpave software predicts pavement performance using material properties,


traffic, pavement temperature, and pavement structure, and displays results in terms of
distress versus time.

5.3 Equipment

The equipment required for preparing and testing compacted specimens is as


follows:

Pan: metal, fiat bottom, for heating aggregates.

Pans: metal, round, approximately 4-1 capacity, for mixing asphalt and aggregates.

Oven and hot plate: electric, for heating aggregates, asphalt, and equipment as
required. Oven must also be capable of maintaining the temperature for short-term
aging.

Scoop: for batching aggregates.

Containers: gill-type tins, beakers, pouring pots, or sauce pans, for heating asphalt.

Thermometers: armored, glass, or dial-type with metal stem, 10C to 230C, for
determining temperature of aggregates, asphalt, and asphalt mixtures.

116
o

o,-_

_ o

_ _o 4c

-_ 0
.,"I

_, o "N

_ "_ _ o

"" 117
= _ cN

.......... 0 0

o
e_ 0

.._ _,

_._ "_ _._ _._ _I_

_ _ '2 ......

0 o_ _ o o 0

o _ o o

_ -_ __ __ _

"_ _

118
Balance: 5-kg capacity, sensitive to 1 g for weighing aggregates and asphalt.
Balance, 2-kg capacity, sensitive to 0.1 g for weighing compacted specimens.

Mixing spoon: large, or trowel, small.

Spatula: large.

Mechanical mixer: (optional) commercial bread dough mixer, 4-1 capacity or larger,
equipped with two metal mixing bowls and two wire stirrers.

Gloves." welders, for handling hot equipment, and Gloves: rubber, for removing
specimens from water bath.

Marking crayons: for identifying test specimens.

SHRP gyratory compactor: capable of providing a consolidation pressure of 0.60


MPa, an angle of gyration of 1.25 degrees, and speed of gyration of 30.0 rpm, as
described in SHRP Method of Test M-002.

Cylindrical molds: large enough to accommodate the following specimen size


requirements: 150 mm diameter, 150 mm maximum height, 90 mm minimum height.

Extrusion jack or arbor press: for extruding compacted specimens from mold.

Superpave shear test device: meeting the requirements presented in SHRP Method of
Test M-003.

Indirect tensile test device: meeting the requirements of SHRP Method of Test M-
005.

5.4 Selecting Mixture Materials for Volumetric Design

The first important step in the level 3 mix design procedure is selecting materials
suited to the demands of traffic and environment expected over time for the paving project.

(1) Select the fine and coarse aggregate and trial blends in accordance with the details
provided in chapter 2 of this manual.

119
(2) Select the performance grade of the asphalt binder in accordance with the details
provided in chapter 2.

(3) Conduct the net adsorption test to evaluate asphalt-aggregate compatibility in


accordance with the details provided in chapter 2.

(4) Conduct initial trial asphalt content and trial aggregate gradation evaluations and
determine the design aggregate gradation and design asphalt content, based on
volumetric property requirements in accordance with the procedures provided in
chapter 3 of this manual. The design traffic level ('Ndesign) discussed in chapter 3 is
greater than 10 7 ESALs for the level 3 mix design. Ninit and Nmax are presented in
table 3.4.

5.5 Selecting Asphalt Contents to Yield 3, 4 and 6 percent Air Voids

The proced_ presented ....in _is....section corresponds...........


_th .........
block 3'2 of figure 5_t, the

(1) Refer to section 3.8.3, for the level 1 volumetric mix design. At the design number
of gyrations (Ndesign)volumetric properties are calculated for each asphalt content as
shown in table 3.9. Air voids, VMA, and VFA are plotted for each asphalt content
as shown in figure 3.9.

(2) From the plot of percent air voids versus percent asphalt content shown in figure
3.9, determine the asphalt contents at three, four and six percent air voids.

(3) Designate the asphalt contents at three, four and six percent air voids as high,
design, and low, respectively.

5.6 Determining Effective Temperature for the Screening Test for


Tertiary Creep

Determine the effective temperature (Teff)at which to conduct the tertiary creep
screening test. Refer to section 4.7.

5.7 Tertiary Creep Mix Evaluation

The procedure outlined in this section corresponds with blocks (3-3 _d 3-4) in fig_
5il, the flow chart for teveI 3 _x design '

120
The tertiary creep evaluation is a screening process to identify a mix that exhibits
tertiary creep (plastic flow) leading to gross mix instability. If the mix fails the screening
test it will be necessary either to make minor adjustments to the mix proportioning or to
redesign the mix completely. If the mixture passes the test requirements, it is then
subjected to the battery of tests for permanent deformation, fatigue, and low-temperature
cracking characterization.

Evaluate the mix's potential for tertiary creep by performing the steps outlined in
section 4.8 of this manual.

5.8 Permanent Deformation and Load-Associated Fatigue Property


Characterization

The procedures outlined in this section correspond with bioekS _3i61 3i71 _d 3_8) in
figure 5,1, the flow ehm for teveI 3 _x designi

This section presents appropriate test methods to determine material properties for use in
predicting the mixture's potential for permanent deformation and load-associ_ited fatigue
over its service life.

(1) Prepare duplicate sets of specimens (150 mm diameter by 50 mm high) compacted


at Ndesign by SHRP Method of Test M-002 to seven percent air voids at the high,
design, and low asphalt contents as determined in section 5.5. The total number of
specimens prepared is 30 (10 specimens per asphalt content times 3 asphalt contents
= 30).
(2) Select twelve of these specimens (4 specimens per asphalt content times 3 asphalt
contents = 12) to test permanent deformation and load-associated fatigue properties.
The remaining eighteen specimens will be used to determine the indirect tensile
strength in step (7) below.

(3) Frequency sweep at constant height test: Test six of the 12 specimens (2 specimens
at three asphalt contents = 6) prepared above by performing the constant height
frequency sweep in accordance with SHRP Standard Practice P-005. Conduct nine
frequency sweeps at 10, 5, 2, 1, 0.5, 0.2, 0.1, 0.05, and 0.02 Hz. Each series of
sweeps, at each asphalt content, is conducted at 4, 20, and 40C, for a total of 27
sweeps. Record axial deformation, shear deformation, axial load, and shear load at a
rate of about 50 data points per load cycle in a format suitable for analysis by the
Superpave software.

(4) Simple shear at constant height: Using the same six specimens tested in step 3,
conduct the constant height simple shear test, in accordance with SHRP Standard

121
Practice P-005. Conduct the test at each asphalt content at 4, 20 and 40C. Record
the axial deformation, the shear deformation, the axial load and the shear load at a
rate of about 10 data points per second in a format suitable for analysis by the
Superpave software.

(5) Volumetric test: Using the remaining six specimens (2 specimens at three asphalt
contents = 6) conduct the volumetric test in accordance with SHRP Standard
Practice P-005. Perform the test for each asphalt content at 4, 20, and 40C.
Record the axial deformation on both sides of the specimens, radial deformation, and
hydrostatic pressure in a format suitable for analysis by the Superpave software.

(6) Uniaxial strain test: Using the same six specimens tested in step 5 conduct the
uniaxial strain test in accordance with the SHRP Standard Practice P-005. Conduct
the test for each asphalt content at 4, 20, and 40C. Record the axial deformation on
both sides of the specimens, radial deformation, axial load, and confining pressure
during the cycles specified in SHRP P-005 and in a format suitable for analysis by
the Superpave software.

(7) Indirect tensile strength test: Using the remaining eighteen specimens (6 specimens
per asphalt content times three asphalt contents = 18) prepared in step (1) conduct
the indirect tensile strength test in accordance with SHRP Method of Test M-005 at
-10, 4, and 20C. Record the maximum peak load at a ram load rate of 50 mm per
minute. The data must be recorded in a format suitable for analysis by the
Superpave software.

5.9 Predicting Permanent Deformation and Load-Associated Fatigue


Cracking at the Pavement Design Life

The procedure outlined in this section corresponds with block (3-10) in figure 5.1, the
flow chart for level 3 mix design.

(1) The data recorded in section 5.8, steps (3) through (7) are entered into the Superpave
software in a format suitable for analysis.

(2) Permanent deformation is predicted by the Superpave software system for a standard
load applied to a pavement structure at the design pavement temperature. Stresses
are calculated within each pavement layer using pavement temperature and mixture
stiffness obtained from the frequency sweep test data. Total permanent deformation
(rut depth) at the pavement surface is calculated versus time.

(3) Fatigue cracking is estimated using a single season analysis to calculate fatigue.
Stresses within each pavement layer are calculated for the layer stiffness

122
representative of the temperature under a standard load. Dissipated energy is
calculated for each season, and the cumulative dissipated energy is used to estimate
percent fatigue cracking. Total area of fatigue cracking as a percentage of total
pavement area is predicted versus time.

5.10 Low-Temperature Cracking (Nonload Associated) Property


Characterization

This section outlines appropriate test methods to determine the material properties
needed to predict the mixture's potential to experience low-temperature cracking.

(1) Prepare nine triplicate sets of specimens compacted by SHRP Method of Test M-002
to 7 percent air voids at the high, design and low asphalt contents as determined in
sections 5.5. Twenty seven specimens are prepared (9 specimens per asphalt content
times 3 asphalt contents = 27).

(2) For each asphalt content, test 3 specimens each at -20C, 3 specimens at -10C, and
3 specimens at 0C by performing the indirect tensile creep and strength tests in
accordance with SHRP Method of Test M-005. For each specimen calculate and
record the creep compliance at 100 seconds in a format suitable for analysis by
Superpave. Increase the load at a rate of 12.5 mm of ram movement per minute
until failure occurs. Record the peak load in a format suitable for analysis by the
Superpave software.

5.11 Predicting Low-Temperature Cracking at the Pavement Design Life

I IIIIIIII I I lllllllllllllllllllll

(1) The data recorded in section 5.10 are entered into the Superpave software in a
format suitable for analysis.

(2) Low-temperature cracking is predicted using low-temperature creep behavior and


low-temperature fracture strengths. Predictive models are used with statistical
weather data to predict stresses caused by overnight temperature cycles. Fracture
mechanics is used in conjunction with tensile strength and predicted tensile stress to

123
determine the advancement of cracking. The test estimates the development of crack
spacing with time.

5.12 Selecting the Optimum Asphalt Content

flow ch_.: of level 3 :mix desi _, ..... ..... .......... ........

The predicted values of permanent deformation (rut depth), fatigue cracking (percent
of area), and low-temperature cracking (crack spacing) are used to select the optimum
asphalt content.

(1) The magnitudes of the predicted values for the distress factors determined in
sections 5.9 and 5.11 are plotted against the asphalt contents tested in sections 5.8
and 5.10 (figure 5.2). Table 5.3 presents suggested acceptable levels of
performance for each distress. These may be adjusted as the design agency acquires
more data or totally replaced by the design agency's specific levels.

Table 5.3. Suggested Acceptable Levels Of Performance Requirements


(Level 3 Mix Design)

Distress Factor Acceptable level

Permanent Deformation
Reliability (%) 95
Average rut depth (mm) 10

Fatigue Cracking
Reliability(%) 95
Percent of area 10

Low-TemperatureCracking
Reliability(%) 80
Crack spacing,m > 150

(2) The design agency's acceptable levels for permanent deformation, fatigue cracking,
and low-temperature cracking are identified on the vertical axis of the graphs and
projected to the right to intersect with the plot. At this point of intersection, a
vertical line is projected perpendicular to the horizontal axis as shown in figures 5.2.

124
Figure 5.2. Selection of an Optimum Asphalt Content Range Satisfying Three
Distress Factor Conditions

15.0 -

12.5 -

(mm) 10.0

7.5 -

RutDepth Accepta__ _,'_1 I

5.5 6.0 6.5


Low Design High
Asphalt Content
(a)
20

15
FatigueCracking
(PercentArea) 10

0 I

5.5 6.0 6.5


Low Design High
Asphalt Content
(b)
0 -

Low-Temperature
Cracking Accepta
(Crackspacing,m)150

300
5.5 6.0 6.5
Low Design High
Asphalt Content
(c)

125
Asphalt contents to the left of the vertical line in the top third of figure 5.2 (as
shown by the shaded area) satisfy the permanent deformation requirement. Asphalt
contents to the right of the dotted of the vertical line in the middle and bottom thirds
of figure 5.2 (as shown by the shaded area) satisfy the fatigue and low-temperature
requirements.

(3) The dashed area indicates the range of asphalt contents that will satisfy all three
distress factor requirements related to the design agency's acceptable levels. Select
an asphalt content within this band and designate it as the optimum asphalt content
to be used for mix production.

(4) If only two distress factors are of prime importance to the designer, a similar
procedure is used as in step 3 (figure 5.3). For this example, permanent deformation
and fatigue cracking are being considered. The dashed area indicates the range of
asphalt contents that will satisfy the design agency's acceptable levels for these
distress types. The optimum asphalt content is selected within this range.

(5) If one distress factor is of primary concem, the following approach is used. Refer to
figure 5.2 or 5.3. Using permanent deformation as an example, the asphalt contents
to the left of the vertical line (as shown by the shaded area) satisfy the permanent
deformation requirement. The same approach is used for fatigue or low-temperature
cracking, except the shaded area to the right of the vertical line as shown in figure
5.3 satisfies the requirement. The optimum asphalt content may be selected within
this range identified by the shaded area.

(6) If a range of acceptable asphalts cannot be established to select an optimum asphalt


content that satisfies the distress factors evaluated, the mix proportioning must be
adjusted or a modifier or modified binder should be considered in the mix design.

(7) The Superpave software provides the comparative plots discussed in steps 2 through
5. The plots are generated from the test as input data and the design agency's
acceptable level of performance requirements.

5.13 Treating Modified Materials and Mixtures

If a modifier is used to enhance the mix performance capability, the procedures


outlined previously for permanent deformation and low-temperature cracking apply also to
modified mixtures. The prediction models used in the Superpave software will evaluate
pavement performance of both conventional and modified asphalt binders and paving mixes
in terms of permanent deformation and low-temperature cracking.

However, if fatigue cracking is the principal distress type addressed by a modifier,


flexural beam testing is recommended in place of the simple shear at constant height,
indirect tensile strength, frequency sweep, and related procedures. No prediction capability

126
Figure 5.3. Selection of an Optimum Asphalt Content Range Satisfying Two Distress
Factor Conditions

15.0 -

12.5 -

Rut Depth
(mm) 10.0 Acceptable ......

7.5 - _-_

_1::_11:
:%1::%'1: I

5.5 6.0 6.5


Low Design High
Asphalt Content
(a)
20-

15-

Fatigue
(Percent Cracking
Area) 10 Acceptab ie_..,.,._
:_

i :

5 -
.... i

0 I
5.5 6.0 6.5
Low Design High
Asphalt Content
(b)

127
is available for this situation with the Superpave models. A comparison of the cycles to
failure can be made between the conventional and modified mixtures.

5.14 Proof Testing

Proof testing provides an independent confirmation of routine Superpave test results


and laboratory performance estimates for situations with severe service requirements; when
an exceptional degree of design reliability is required; when unusual or new materials are
used; or when paving mixes have a top-size aggregate greater than two inches.

5.14.1 Rolling Wheel Compaction

This compaction procedure produces a slab from which beams may be sawn or large
cylindrical specimens cored. It produces test specimens with a known air void content at a
specified asphalt content. The procedure recommended is SHRP Method of Test M-008,
Preparation of Asphalt Concrete Mixture Test Specimens by Means of Rolling Wheel
Compaction.

5.14.2 Wheel-Tracking Device

Wheel-tracking test devices of various sizes and degrees of complexity are available
to evaluate a mixture's resistance to permanent deformation. These devices employ
pneumatic rubber-tired, steel wheels or hard-rubber-tired wheels. The equipment measures
the rut depth created by a repeated passage of a wheel over a slab prepared in the
laboratory or cut from a field pavement.

The dependence of the rut depth in the specimen on material types, number of
loading cycles, tire pressure, wheel load, temperatures, and moisture conditioning can be
determined. The test is a laboratory simulation of the rutting phenomenon with actual
pavement stress conditions. One device that has been used successfully for extensive mix
design evaluation is the rolling wheel rut tester developed at the Laboratoire Central des
Ponts et Chausees (LCPC) in Nantes, France.

5.14. 3 Flexural Beam Fatigue Test

The design agency can employ this test in level 3 mix design to evaluate modified
mixtures discussed in section 5.12. The test is conducted in accordance with SHRP
Method of Test M-009, Determining the Fatigue Life of Modified and Unmodified Hot Mix
Asphalt Subjected to Repeated Flexural Bending. The beams are sawn from a rolling wheel
compacted slab. The failure point is defined as the load cycle at which the specimen
exhibits a 50 percent reduction in stiffness relative to the initial stiffness. No prediction

128
based on the Superpave prediction models can be made. Instead a comparison of the cycles
to failure is made among different mix designs or materials.

5.14.4 Thermal Stress Restrained Specimen Test

The design agency may opt to verify that the fracture temperature established in the low-
temperature cracking analysis is realistic or reasonable. The measured fracture temperature
may be compared to historical low pavement temperatures related to low-temperature
cracking. The test to determine the fracture strength and fracture temperature for
verification purposes is SHRP Method of Test M-010, Determining the Fracture Strength
and Temperature of Modified and Unmodified Hot Mix Asphalt Subjected to Cold
Temperatures. The test system is capable of cooling the mixture specimen at a constant
rate while restraining the specimen from contraction; it periodically measures the tensile
load and the specimen temperature from the beginning of the test to specimen failure. No
prediction based on the Superpave prediction models can be made. The fracture strength
and temperature are evaluated only by comparative analysis.

129
6

Superpave Mix Design: An Example


This chapter provides a worked example of a Superpave level 1/level 3 mix design for a
simulated paving project located on 1-43 near Milwaukee, Wisconsin.

6.1 Select Materials

The first step in a Superpave mix design is selection of the asphalt binder and
aggregate stockpiles that meet the environment and traffic requirements present at the
paving project.

The design equivalent single axle loads (ESALs) for the project are determined to be
18 million. This puts the design in the category of 10 million to 30 million ESALs, or
Traffic level 5. Traffic levels are used to determine design requirements such as number of
design gyrations for compaction, aggregate physical property requirements, and mixture
volumetric requirements. The traffic level also determines the level of mixture design
required. For traffic levels of 10 7 ESALs and higher, a level 3 design is recommended.
This is a full level I design to determine mixture volumetric properties, followed by
performance prediction tests.

The mixture in this example is an intermediate course mixture. It will have a


nominal maximum particle size of 19.0 mm. It will be placed at a depth less than 100 mm
from the surface of the pavement.

Environmental conditions are determined from weather station data stored in the
Superpave software database. The project is near Milwaukee, which has two weather
stations; the relevant data are presented in table 6.1.

131
Table 6.1. Project Weather Data

Weather Station Min. Print. Max. Pvmt. Binder Grade Design Air
Temp. (C) Temp. (C) Temp. (C)
Low Reliability (50%)

Milwaukee Mt. Mary -26 52 PG 52-28 32


Milwaukee WSO AP -25 51 PG 52-28 31

Paving location -26 52 PG 52-28 32

High Reliability (98%)

Milwaukee Mt. Mary -32 55 PG 58-34 36


Milwaukee WSO AP -33 54 PG 58-34 34

Paving location -32 55 PG 58-34 35

The low and high reliabilities are in parentheses. These are the probabilities that in a
given year the actual pavement temperature will not exceed or fall below the minimum and
maximum pavement temperatures, respectively, listed in table 6.1. It is possible to select
high reliability for one condition, such as minimum pavement temperature, while selecting
low reliability for another condition, such as maximum pavement temperature. In this
example, the designer chooses high reliability for all conditions. This requires a PG 58-34
binder) The average design high air temperature is 35C.

An appropriate asphalt binder is selected and tested for specification compliance.


Test results are presented in table 6.2.

Comparing the test results to the specification criteria, the mix designer verifies that
the asphalt binder meets the required PG 58-34 grade. Specification testing requires only
that rotational viscosity be performed at 135C. Additional testing is performed at 175C
to establish mixing and compaction temperatures. Based on the test results, 165 to 173C
is determined as the mixing range, and 150 to 157C as the compaction range.

Next, the designer selects the aggregates to use in the paving mix. For this example,
there are five stockpiles of materials consisting of three coarse materials and two fine
materials. The materials are split into representative samples, and washed sieve analysis is
performed for each aggregate. Test results are presented in section 6.2.

t In practice, additional grading requirements based upon traffic speed and volume should
be considered; see chapter 2.

132
Table 6.2. Binder Specification Test Results

Test Property Test Result Criteria

Original Binder

Flash point n/a 304C 230C minimum


Rotationalviscosity 135C 0.575 Pa-s 3 Pa-s maximum
Rotational viscosity 175C 0.142 Pa-s n/a
Dynamic shear rheometer G*/sin5 @ 58C 1.42 kPa 1.0 kPa minimum
RTFO-aged Binder
Mass loss n/a -0.14% + 1.0% maximum
Dynamic shear rheometer G'/sin 5 @ 58C 2.41 kPa 2.2 kPa minium

PAV-aged Binder

Dynamic shear rheometer G'sin _5@ 16C 1543 kPa 5000 kPa max.
Bending beam rheometer Stiffness @ -24C 172,000 kPa 300,000 kPa max.
Bending beam rheometer Slope m @ -24C 0.321 0.300 min.

The specific gravity (bulk and apparent) of each aggregate is determined. These
specific gravities are used in trial binder content calculations and the percent VMA
calculations. Test results are presented in table 6.3:

Table 6.3. Aggregate Specific Gravities

Aggregate Bulk Specific Gravity Apparent Specific Gravity

#1 stone 2.703 2.785

1/2" chip 2.689 2.776

3/8" chip 2.723 2.797


Manufactured sand 2.694 2.744

Screen sand 2.679 2.731

In addition to sieve analysis and specific gravity determination, the Superpave design
method requires the performance of certain consensus aggregate tests to ensure that the
aggregate blend selected for the mix design is acceptable. Four tests are required: coarse
aggregate angularity; fine aggregate angularity; thin and elongated particle determination;
and clay content. In addition, the specifying agency can select any other aggregate tests
deemed important. These tests may include sulfate soundness, Los Angeles abrasion, and
friable particle content among others.

133
At this point the designer has the option of performing the four consensus aggregate
tests on each stockpile or performing the test on each trial aggregate blend. The former
choice allows the designer an opportunity to use the test results in narrowing the blend
percentages of the aggregates for mix design. It also allows for greater flexibility if
multiple trial blends are attempted. Performing the tests on each trial blend is technically
correct, but requires that aggregate blending be performed ahead of the determination of the
aggregate physical characteristics. Either method is appropriate. For this example, the test
results are shown for each stockpile as well as for the aggregate trial blends.

6.1.1 Coarse Aggregate Angularity

This test is performed on the coarse aggregate particles of the aggregate stockpiles.
The coarse aggregate particles are defined in the test as particles larger than 4.75 mm. Test
results are presented in table 6.4.

Table 6.4. Coarse Aggregate Angularity

Aggregate l+Fractured Faces, % Criterion 2+Fractured Faces, % Criterion


#1 stone 92 88
I/2" chip 97 95% rain 94 90% rain
3/8" chip 99 95

6.1.2 Fine Aggregate Angularity

This test is performed on the fine aggregate particles of the aggregate stockpiles.
Particles smaller than 2.36 mm are defined as fine aggregate particles. Test results are
shown in table 6.5.

Table 6.5. Fine Aggregate Angularity

Aggregate % Air Voids (Loose) Criterion (%)


Manufactured Sand 62 45 rain
Screen Sand 35

Note that this test is not performed on the three coarse aggregates, even though they
have some small percentage of particles passing the 2.36 mm sieve. The #1 stone has 1.9
percent passing, the 1/2" chip has 2.6 percent passing, and the 3/8" chip has 3.0 percent
passing the 2.36 mm sieve.

Table 6.5 gives the criterion for fine aggregate angularity based on the traffic (level
5) and the depth from the surface (<100 mm). The criteria change as the traffic level and
layer depth change. The criteria are also based on the test results from the aggregate blend
rather than individual materials. The screen sand appears to be below the minimum

134
criterion, but it can be used as long as the selected blend of aggregates meets the criterion
in table 6.5.

6.1.3 Thin and Elongated Particles

This test is performed on the coarse aggregate particles of the aggregate stockpiles.
For this test the coarse aggregate particles are defined as particles larger than 4.75 mm.
Test results are presented in table 6.6.

Table 6.6. Thin and Elongated Particles

Aggregate % Thin and Elongated Criterion (%)


#1 stone 0
1/2" chip 0 l0 max.
3/8" chip 0

Note that this test is not performed on the two fine aggregates, even though they
have some small percentage retained on the 4.75 mm sieve. The manufactured sand has 4.5
percent retained and the screen sand has 10.5 percent retained on the 4.75 mm sieve.

Table 6.6 also indicates the criterion for the percentage of thin and elongated
particles based on traffic (level 5) only. The criterion changes as the traffic level changes.
The criteria are also based on the test results from the aggregate blend rather than
individual materials. In this case, the aggregates are cubical and not in danger of failing the
criteria.

6.1.4 Clay Content (Sand EquivalenO

This test is performed on the fine aggregate particles of the aggregate stockpiles.
The fine aggregate particles are defined as particles smaller than 2.36 mm. Test results are
shown in table 6.7.

Table 6.7. Clay Content (Sand Equivalent)

Aggregate Air Voids (Loose) (%) Criteria (%)


Manufactured Sand 47 45 min.
Screen Sand 70

Note that this test is not performed on the three coarse aggregates, even though they
have some small percentage passing the 2.36 mm sieve. The #1 stone has 1.9 percent
passing, the 1/2" chip has 2.6 percent passing, and the 3/8" chip has 3.0 percent passing the
2.36 mm sieve.

135
Table 6.7 gives the criterion for clay content (sand equivalent) based on traffic (level
5) only. The criterion changes as the traffic level changes. The criteria are also intended
for application to the test results from the aggregate blend rather than individual materials.
Both fine aggregates are above the minimum requirements, so there the blend should not
present problems.

The material selection process is complete with the completion of the aggregate
testing and selection. The next step in the design is to determine a design aggregate
structure.

6.2 Select Design Aggregate Structure

To select the design aggregate structure, the designer establishes trial blends by
mathematically combining the gradations of the individual materials into a single gradation.
The blend gradation is then compared to the specification requirements for the appropriate
sieves. Gradation control is based on four control sieves: the maximum sieve size (defined
in chapter 2), the nominal maximum sieve (defined in chapter 2), the 12.5 mm sieve, the
2.36 mm sieve, and the 0.075 mm sieve. There is also a restricted zone that is an area on
either side of the maximum density line generally starting at the 2.36 mm sieve and
extending to the 0.3 mm sieve. The minimum and maximum values required for the
control sieves change (as does the restricted zone) as the nominal maximum size of the
mixture changes. Table 6.8 indicates the gradation requirements for this example.

Table 6.8. Gradation Criteria for 19.0 mm Nominal Maximum Size Aggregate

Sieve Minimum (% pass) Maximum (% pass)

25.4 mm 100.0 100.0


19.0 mm 90.0 100.0
12.5 mm - 90.0
2.36 mm 23.0 49.0
0.075 mm 2.0 8.0

Restricted Zone

2.36 mm 34.6 34.6


I. 18 mm 22.3 28.3
0.6 mm 16.7 20.7
0.3 mm 13.7 13.7

Any trial blend gradation has to pass between the control points established on the
five sieves. In addition, it has to be outside of the area bounded by the limits set for the
restricted zone (see figure 6.1).

Any number of trial blends can be attempted, but three is the standard number of
blends. Trial blending consists of varying stockpile percentages of each aggregate to obtain

136
a blend gradation meeting the gradation requirements for that particular mixture. In this
example, three trial blends are attempted: an intermediate blend, a coarse blend, and a fine
blend.

The intermediate blend is combined to produce a gradation that is not close to either
the gradation limits for the control sieves or to the restricted zone. The stockpile
percentages and combined gradation for the intermediate blend (labelled trial blend 1) are
shown in table 6.9.

The coarse blend is combined to produce a gradation that is close to the minimum
criteria for the nominal maximum sieve, the 2.36 mm sieve, and the 0.075 mm sieve. The
stockpile percentages and combined gradation for the coarse blend (labelled trial blend 2)
are presented in table 6.9.

The fine blend is combined to produce a gradation that is close to the maximum
criteria for the nominal maximum sieve and to the restricted zone. The stockpile
percentages and combined gradation for the fine blend (labelled trial blend 3) are indicated
in table 6.9.

All three of the trial blends are shown graphically in figure 6.2. All three trial
blends pass below the restricted zone. This is not necessarily a requirement. It is possible
to have a trial gradation that passes above the restricted zone.

Once the trial blends are selected, a preliminary determination of the aggregate blend
properties is necessary. These can be estimated mathematically from the aggregate
properties determined in tables 6.3 to 6.7. Estimated values are shown in table 6.10.

Table 6.9. Estimated Aggregate Blend Properties

Property Criteria Trial Blend 1 Trial Blend 2 Trial Blend 3


Coarse Ang. 95%/90%min. 96%/92% 95%/92% 97%/93%
Fine Ang. 45% min. 48% 50% 54%
Thin/Elongated 10%max. 0% 0% 0%
Clay Content 45% min. 59% 58% 54%
Combined Gsb n.a. 2.699 2.697 2.701
Combined Gsa n.a. 2.768 2.769 2.767

137
138
_ I_ _ t'_ ,_2 t'N _ _ '_"

_ _. _ _ t",- ,..... _ _ _ _'_ _ ""2.

r_


,.-. (.,,q _

_ _ o_
m m _ 139
Figure 6.2. Aggregate Gradations of Three Trial Blends

19.0ram NOMINAL MAXIMUM SIZE MIXTURE


I00.0
90.0 ""
80.0

70.0 TRIAL BLEND


60.0
50.0 a ....
z 40.0
_) 30.0
a. 20.0
I0.0

0.0
.075 236 12.5 19.0 25.0
SIEVE SIZE (RAISED TO 0.45 POWER),mm

140
Values for coarse aggregate angularity are shown as the percentage of one or more
fractured faces followed by the percentage of two or more fractured faces.

Based on the estimated properties, all three blends appear acceptable. When the
design aggregate structure is selected, the blend aggregate properties will be verified by
testing.

The next step is to evaluate the trial blends by compacting specimens and
determining the volumetric properties of each blend. A minimum of two specimens per
blend is compacted using the SHRP gyratory compactor.

The trial asphalt binder content is determined for each trial blend by estimating the
effective specific gravity of the blend and using the calculations given in chapter 3.

The effective specific gravity, Gsc, of the blend is estimated from its bulk (Gsb) and
apparent (Gs_) specific gravities by using the following equation:
Gse=Gsb +O .8 (Gsa-Gsb) (6-i)

The multiplier, 0.8, can be changed at the discretion of the designer. Absorptive
aggregates may require values closer to 0.6 or 0.5. The calculation for each trial blend is
shown below:

Trial Blend 1: Gsc = 2.699 + 0.8(2.768 - 2.699) = 2.754


Trial Blend 2: G_, = 2.697 + 0.8(2.769 - 2.697) = 2.755
Trial Blend 3: G_ = 2.701 + 0.8(2.767 - 2.701) = 2.754

The volume of asphalt binder absorbed into the aggregate is estimated using the
following equations:

Vba=Ws (i/Gsb-i/Gse) (6 -2 )

(1-va)
Ws:PsX
Pb
--+
Ps (6 - 3 )
Gb Gse

where W_ = weight of aggregate,


Pb = percent of binder (assumed 0.05),
Ps = percent of aggregate (assumed 0.95),
Gb = specific gravity of binder (assumed 1.02), and
V, = volume of air voids (assumed 0.04).

141
Using these equations the following values of Vba are calculated for each trial blend:

Blend 1:

0.95(1-0.04)
095 1 1 (6-4)
017
Vba: [ 1.
0 0205 ] +[ (2_-7--5-4)( 2 699 --
2.754 )] =0

Blend 2:

0.95 (i-0.04)

0.05 ]+[ 0.95 1 1 )]=0 018 (6-5)


Vba=[ 1.02 2.75----_ 2.697 5-----_
2.7 "

Blend 3"

0.95 (i-0.04)

0.05 ] +[ 0.95 1 1 )] =0 016 (6-6)


Vba=[ 1.02 2.755 2.697 2.75---------5
"

The volume of the effective binder can be determined from the equation below:

Vbe:O. 091-0. 029311n (0. 039S n) (6-7)

where S n = the nominal maximum sieve size of the aggregate blend (in mm).

Using this equation the value of Vbe for trial blends 1-3 is:

Vbe=O.091-0.029311n(19.0xO.039) =0.099 (6-8)

Finally, the initial trial asphalt binder content is calculated from the following equation:

Gb(Vbe+Vba) (6-9)
Pbi- (Gb(Vbe+Vba) ) +W s

142
This equation yields a value of 4.4 percent for Pbi for trial blends 1, 2 and 3.

Next, a minimum of two specimens for each trial blend are compacted using the
SHRP gyratory compactor. Two additional specimens are prepared for determining the
mixture's maximum theoretical specific gravity. An aggregate weight of 4500 grams is
usually sufficient for the compacted specimens. An aggregate weight of 1500 grams is
usually sufficient for the specimens used to determine maximum theoretical specific gravity
(Gram).

Specimens are mixed at the appropriate mixing temperature which is 165C to


173C for the selected PG 58-34 binder. The specimens are then short-term aged 2 by
placing the loose mix in a fiat pan in a forced draft oven at 135C for 4 hours. The
specimens are then brought within the compaction temperature range (150C to 157C) by
placing them in another oven for a short time (generally less than 30 minutes). Finally, the
loose mix is removed and either compacted or allowed to cool loose (for maximum
theoretical specific gravity determination).

The number of gyrations used for compaction is determined based on the design
high air temperature of the paving location (35C) and the traffic (level of >10 7 ESALs).
For these two conditions, the number of gyrations are identified as (see table 3.4):

Ninit = 8 gyrations;
Ndesig n = 109 gyrations; and
Nmax = 174 gyrations.

Each specimen is compacted to the maximum number of gyrations, with data


collected during the compaction process. During compaction, the height of the specimen is
continually monitored. Knowing the initial weight of the mix, the fixed volume of the
mold, and the measured height, the density can be continually monitored.

After compaction, the final density of the specimens are determined by AASHTO T
166. The Gmm of each blend is also determined. From this data, the air voids of the
specimen at each gyration can be determined. Finally, the final calculated density of the
compacted specimens are corrected to match the final measured density. The density at
each gyration is subsequently adjusted. There is generally a difference of one to four
percent between the calculated density and the measured density.

2 All specimens of paving mix compacted in the Superpave mix design method are
conditioned as loose mix according to the short-term aging procedure in SHRP Standard
Method of Test M-007.

143
The densification curves are plotted for each trial blend. The percent of the
maximum theoretical density is determined for Ninit (8 gyrations), Ndesign (109 gyrations),
and Nmax (174 gyrations) for each compacted specimen of each trial blend. The air voids
content and the VMA are estimated (see table 6.11). An estimated asphalt binder content to
achieve 4% air voids (96% Gmm ) is determined for each trial blend. The mixture properties
are then estimated at this asphalt binder content (see table 6.12). Estimated properties are
compared to the Superpave level 1 (volumetric) criteria (see table 6.13). For the design
traffic and nominal maximum particle size, the volumetric criteria are as follows:

% Air Voids 4.0


% VMA <13.0 (19.0 mm nominal maximum size mix)
% VFA 65 - 75

There are also criteria for densification that are constant for values at Ninit (8

gyrations), Ndesign (109 gyrations), and Nmax (174 gyrations). They are as follows:

% Gmmat Ninit(8 gyrations) < 89%


o_ Gmm at Ndesig
n (109 gyrations) 96%
o_ Gmm at Nmax (174 gyrations) < 98%

Table 6.11. Summary of Trial Blends

Blend % AC % Gram @ N=8 % Gram@N=174 % Gmm @N=I09 % Air % VMA


Voids

1 4.4 87.1 97.6 96.2 3.8 12.7


2 4.4 85.6 97.4 95.7 4.3 13.0
3 4.4 86.3 96.5 95.1 4.9 13.5

Table 6.12. Estimated Values for Trial Blends

Blend Trial Est. % Gmm o_ Gmm @ o_ Air % %


% AC % AC @ N=8 N=174 Voids VMA VFA

1 4.4 4.3 86.9 97.4 4.0 12.7 68.5


2 4.4 4.5 86.0 97.7 4.0 13.0 69.1
3 4.4 4.7 87.2 97.4 4.0 13.4 70.1

144
Table 6.13. Mixture Criteria

% Grnm at N = 8 gyrations < 89%

% Gram at N = 174 gyrations < 98%


% Air Voids 4.0%

% VMA (19.0 mm nominal maximum size _ 13.0%


mix)

% VFA (< 107ESALs design traffic) 65 - 75%

Finally, the dust-to-asphalt binder ratio criterion is considered. This criterion is


constant for all levels of traffic. It is calculated as the percent by weight of the material
passing the 0.075 mm sieve (by wet sieve analysis) divided by the effective asphalt binder
content (expressed as percent by weight of mix). The dust-to-asphalt ratio must be between
0.6 and 1.2. Table 6.15 presents the results for the three trial blends.

Table 6.14. Dust-to-Asphalt Ratio of Trial Blends

Blend Dust/Asphalt Ratio Criteria


Trial Blend 1 0.84
Trial Blend 2 0.81 0.6 - 1.2
Trial Blend 3 0.95

After establishing all the estimated mixture properties, the data for the three trial
blends are reviewed and a decision is made if one or more are acceptable, or if further trial
blends require evaluation.

Table 6.15. Comparison of Trials Blends

Property Blends That Meet Criterion

%Gram@ N= 8 gyrations 1, 2, 3

%Gram
@ N= 174 gyrations 1, 2, 3
% Air Voids I, 2, 3

% VMA 2, 3

% VFA 1, 2, 3

Trial blend 1 is unacceptable based on its failure to meet the minimum % VMA
criterion. Trial blend 2 is acceptable, but the % VMA is at the lower limit. Trial blend 3
has an acceptable % VMA and meets the criteria for % VFA, dust-to-asphalt ratio, and the
densification criteria. From these data, trial blend 3 is selected as the design aggregate
structure.

145
6.3 Select Design Asphalt Binder Content

Selecting the design aggregate structure is the difficult step. Once the structure is
selected (Trial Blend 3), specimens are compacted at several asphalt binder contents. The
mixture properties are then evaluated with respect to the asphalt binder content.

A minimum of two specimens are compacted 3 at each of the following asphalt


contents: estimated asphalt binder content, the estimated asphalt binder content +0.5%, and
the estimated asphalt binder content +1.0%. For trial blend 3, the binder contents for the
mix design are thus 4.2%, 4.7%, 5.2%, and 5.7%. A minimum of four asphalt binder
contents are evaluated. A minimum of two specimens are also prepared for determination
of the maximum theoretical specific gravity. Specimens are prepared and tested in the
same manner as the specimens in section 6.2.

Mixture properties are evaluated for the selected blend at the different asphalt binder
contents. From these data points, plots of % air voids, % VMA, % VFA and density versus
asphalt binder content are prepared. The design asphalt binder content is established at 4%
air voids. All other mixture properties are checked at that asphalt binder content to verify
that they meet the volumetric criteria. The design values for a 19.0 mm nominal maximum
size mix (trial blend 3) are presented in table 6.16:

Table 6.16. Design Mix Properties @ 4.7 % Asphalt Content

Mix Property Result Criteria

% Air Voids 4.0 4.0


% VMA 13.2 13.0 rain.
% VFA 70.0 65 - 75
Dust-to-Asphalt Ratio 0.97 0.6- 1.2
% Gram @ Ninit = 8 87.1 < 89
,/o Gram @ Nma x = 174 97.5 < 98

6.4 Evaluate Moisture Sensitivity

The final step in the level 1 mix design is to evaluate the moisture sensitivity of the
design mixture. This step is accomplished by performing AASHTO Method of Test T 283
on the design aggregate structure at the design asphalt binder content. Specimens are
compacted to approximately 7 percent air voids. One subset of three specimens is the
control subset. The other subset of three specimens is the conditioned subset. The

3All specimens of paving mix compacted in the Superpave mix design method are
conditioned as loose mix according to the short-term aging procedure in SHRP Standard
Method of Test M-007.

146
conditioned subset is subjected to partial vacuum saturation followed by an optional freeze
cycle, followed by a 24 hour thaw cycle at 60C. All specimens are tested to determine
their indirect tensile strengths. The moisture sensitivity is determined as a ratio of the
tensile strengths of the conditioned subset divided by the tensile strengths of the dry
(control) subset (table 6.17). The minimum criterion is 80 percent retained tensile strength.
The design blend met the minimum requirement with an 83 percent retained tensile strength
ratio.

Table 6.17. Moisture Susceptibility of Design Asphalt Blend


by AASHTO T 283

SampleNumber 1 2 3 4 5 ]6
Dry Strength(kPa) 890.1 859.0 870.8
Wet Strength(kPa) 713.5 704.5 745.2

Average Dry Strength:873.3 kPa


Average Wet Strength:721.1 kPa
TensileStrengthRatio: 83%

6.5 Superpave Level 3 Mixture Design Example

6.5.1 Introduction

The following example describes the use of the Superpave accelerated performance
tests and performance models to support the level 3 mix design described in chapter 5 of
this manual. Only one asphalt content is considered for discussion purposes. The examples
consider only permanent deformation. Similar examples could be shown for fatigue
cracking and low-temperature cracking with the Superpave pavement performance model.

The Superpave pavement performance model is composed of four parts (Figure 6.3 )
as follows:

Mixture characterization or Material Property Model;


Environmental Effects Model;
Pavement Response Model; and
Pavement Distress Model.

147
Figure 6.3. Flow Diagram of the Different Parts of the Superpave Pavement
Performance Model

PROGRAM INPUTS MODULE


SUBGRADE SUPPORT-LAYER THICKNESS-
TRAFFIC- TEMPERATURES-RAINFALL

o- . * - =.*..eeoe,e,e,,e ee!eeeee *u ,eeeeeeeelee,e ee eeee,,ee*_

V T
ENVIRONMEN RESPONSE DISTRESS OUTPUT
EFFECTS _ MODULE
MODEL l [ MODEL MODEL AMOUNT
DISTRESS
OF

PROPERTY
MODEL
L_ MATERIAL t_
9

VOLUMETRIC UNIAXlAL SIMPLE SHEAR INDIRECT INDIRECT


COMPRESSION-COMPRESSION- SHEAR- FREQUENCY-TENSILE - TENSILE
SWEEP CREEP STRENGTH

PERFORMANCE BASED LABORATORY TESTS MODULE

148
The mixture characterization program calculates the non-linear elastic, viscoelastic,
plastic and fracture properties of a mixture from the performance-based laboratory tests.
These properties are used with the pavement response program to evaluate the behavior of a
mixture subjected to traffic and/or environmental loads. Some of these material properties
are also used directly in the pavement distress model. The pavement response program,
which is a finite element program, calculates the stresses and strains in the asphalt-
aggregate mixture from wheel loads and environment loads. The pavement distress
program takes the relevant mixture properties and the appropriate stresses and strains
calculated with the finite element program and calculates the amount of cracking (from
wheel loads and environmental loads) and rutting with time.

6.5.2 Level 3 Mix Design and Pavement Performance Prediction Approach

The general steps to performing a level 3 mix design are as follows:

(1) Perform level 1 mix design as described in chapter 3 and illustrated in the level 1
example problem of this chapter. The output from the design includes the
following:

selected asphalt binder grade;


design asphalt binder content;
selected aggregate materials; and
design aggregate percentages.

(2) Define project where mix is to be used:


design life;
expected traffic;
select climatic region;
select weather files;
select design level;
define pavement cross section;
select seasonal adjustment factors; and
select asphalt contents for analysis.

(3) Perform repeated shear screening test for tertiary creep.

(4) Perform performance based tests.

(5) Run performance prediction models.

(6) Determine acceptability and select job mix.

149
6.5.2.1 Define Project Where Mixture is to be Used

The design life for the project is specified to determine the number of years of
weather data required to drive the performance models. Design traffic is defined as the
expected number of ESALs which will be applied over the pavement life. Within the
performance models the design number of ESALs is divided by the design life to obtain the
number of axle passes which will be stimulated by the performance models. The
performance models do not escalate annual traffic to simulate traffic growth rate.

Climatic region of the project is selected using the Environmental Effects Model
map shown in figure 6.4. Associated with each of the nine environmental effects regions
are default parameter files for sunshine, rainfall, wind, etc. which will be used in the
environmental model.

Temperature files which will be used to drive the environmental effects program are
selected. A data base of weather data is available with the Superpave software. These
weather stations have ten years of historical daily minimum/maximum temperatures.

Next, the desired level of design is selected. Independent selections can be made
with Superpave for low temperature cracking and permanent deformation/fatigue cracking.
Options include level 3, level 2, or level 1. In the example that follows, permanent
deformation and the level 3 mix design are considered for illustrative purposes.

The pavement cross section in which the mixture is to be located must be defined.
The entire cross section can include up to seven pavement layers including subgrade. A
maximum of two new asphalt layers may be used.

Seasonal adjustment factors can be selected for traffic and lower layer moduli. The
designer can define month-to-month variations in traffic level if desired. Lower layers with
moduli affected by thaws weakening, desiccation, etc. can be simulated by using monthly
adjustment factors.

6.5.2.2 Select Asphalt Contents for Analysis

Performance prediction can be done for a range of asphalt contents. For each mix
design in the pavement cross section asphalt content can be selected as low, medium or
high. Medium asphalt content is the percent asphalt binder obtained from the level 1 mix
design. High asphalt content is the percent asphalt binder which produces 3% air voids at
the design number of gyrations. Low asphalt content is the percent which produces 6% air
voids.

150
!..,

-_ 151
The number of selected asphalt contents has a direct impact on the amount of
required testing and a multiplying effect on the amount of computer analysis time. If two
new asphalt layers are used, each with three asphalt contents there is a total of six
combinations which must be analyzed for performance prediction.

6.5.2.3 Perform Repeated Shear Screening Test

The repeated shear at constant stress ratio screening test is performed as discussed in
sections 5.6 and 5.7 at the highest selected asphalt content. Mixtures which fail are
excluded from performance based testing.

6. 5.3 Performance-Based Tests and Results

The tests discussed below are required for the prediction of permanent deformation
in the level 3 design. The performance-based tests are as follows:

Volumetric;
Uniaxial Strain;
Simple Shear at Constant Height; and
Frequency Sweep at Constant Height.

The procedures presented in this section determine material properties by appropriate


test methods for use in predicting the mixture's potential for permanent deformation.

(1) Prepare duplicate sets of specimens (150mm diameter by 50mm high)


compacted at Ndes_g" by SHRP Method of Test M-002 to 7% air voids at the
high, design and low asphalt contents. The total number of specimens
prepared for permanent deformation testing is 12 (4 specimens per asphalt
content times 3 asphalt contests = 12).

Note - All specimens of paving mix compacted in the Superpave mix design method are
conditioned as loose mix according to the short-term aging procedure in SHRP Standard
Method of Test M-007.

(2) Frequency Sweep at Constant Height: Six of the twelve specimens (2


specimens at three asphalt contents = 6) prepared above are tested by
performing the constant height frequency sweep in accordance with SHRP
Standard Practice P-005. Nine frequency sweeps are conducted at 10, 5, 2, 1,
0.5, 0.2, 0.1, 0.05, and 0.02 Hz. Each series of sweeps, at each asphalt
content, is conducted at 4C, 20C and 40C for a total of 27 sweeps. Axial
deformation, shear deformation, axial load and shear load is recorded at a rate
of about 50 data points per load cycle in a format suitable for analysis by the
Superpave software. Table 6.18 is an illustration of the data output generated

152
from this test for one asphalt content.

(3) Simple Shear at Constant Height: On the same six specimens tested in the
frequency sweep procedure, conduct the simple shear test at constant height
in accordance with SHRP Standard Practice P-005. The test is conducted at
each asphalt content at 4C, 20C and 40C. After preconditioning, the shear
stress on the specimen is increased at a rate of 70 kPa/s to a level determined
by the test temperature. The shear stress is maintained for 10 seconds while
the compressive load is controlled by feedback to maintain the specimen at
constant height. The shear stress is then reduced at a rate of 25 kPa/s to 0
kPa and held for an additional 10 seconds. The axial deformation, shear
deformation, axial load and shear are recorded at a rate of 10 data points per
second in a format suitable for analysis by the Superpave software. Table
6.19 presents an example of a portion of the test output developed from this
test for one asphalt content.

(4) Volumetric Test: On the remaining six specimens (2 specimens at three


asphalt contents = 6) prepared above, conduct the volumetric test in
accordance with SHRP Standard Practice P-005. Perform the test for each
asphalt content at 4C, 20C and 40C. The axial deformation on both sides
of the specimens, the radial deformation and the hydrostatic (confining)
pressure are recorded in a format suitable for analysis by the Superpave
software. Table 6.20 presents an example of the data output from this test for
one asphalt content.

153
Cl

"a x

t_

154
Table 6.19. Example Test Data From Simple Shear at Constant Height Test

Time Axial Load Shear Load LVDT #1 LVDT #1


(see) (N) (N) Axial Shear
Deform_ion Deform_ion
(mm) (mm)

0 34.27 5.23 -0.0290 0.5016


0.1 32.11 124.41 -0.0291 0.5016
0.2 27.79 243.58 -0.0290 0.5016
0.3 25.64 368.16 -0.0291 0.5020
0.4 23.48 481.92 -0.0290 0.5013
0.5 21.32 601.09 -0.0291 0.5010
0.6 21.32 736.51 -0.0291 0.5010
0.7 23.48 850.26 -0.0290 0.5010
0.8 23.48 969.43 -0.0290 0.5013
0.9 21.32 1094.02 -0.0291 0.5010
1 19.16 1213.19 -0.0289 0.5007
1.1 21.32 1337.77 -0.0290 0.5010
1.2 21.32 1467.78 -0.0291 0.5007
1.3 23.48 1586.94 -0.0291 0.5007
1.4 23.48 1706.11 -0.0289 0.5004
1.5 23.48 1836.12 -0.0292 0.5007

(5) Uniaxial Strain Test: On the same six specimens tested the volumetric test,
conduct the uniaxial strain test in accordance with the SHRP Standard
Practice P-005. The test is conducted for each asphalt content at 4C, 20C
and 40C. The specimen is covered around its circumference with a close-
fitting rubber membrane. After a preconditioning step, the shear stress
applied to the specimen is increased at a rate of 70 kPaJs to a level
determined by the test temperature. The shear stress is held for 10 seconds
and then reduced to 7 kPa/s to a level of 25 kPa where it is held for 30
seconds. During this cycle of shear stress, the confining pressure is adjusted
through a feedback loop to maintain the circumference of the specimen at a
constant value. The axial deformation on both sides of the specimens, the
radial deformation, the axial load and the confining pressure are recorded at a
rate of 10 data points per second in a format suitable for analysis by the
Superpave software. Table 6.21 presents data output from this test for one
asphalt content.

155
Table 6.20. Example Test Data from the Volumetric Test

Time Sequence LVDT #2 Axial LVDT #4 Confining Radial


(sec) Deform_ion Axial P_ssure Deform_ion
(mm) De_rm_ion (kP_ (mm)
(mm)
0 I 0.08942 -0.30270 13.11 -0.38115
0 2 0.10634 -0.29967 27.60 -0.35571
0.088 2 0.10665 -0.29967 28.29 -0.35515
0.176 2 0.10757 -0.29899 32.43 -0.35405
0.264 2 0.10880 -0.29866 37.26 -0.35294
0.352 2 0.11003 -0.29764 42.78 -0.35073
0.44 2 0.11157 -0.29764 47.61 -0.34852
0.528 2 0.11341 -0.29697 53.82 -0.34630
0.616 2 0.11495 -0.29596 59.34 -0.34298
0.704 2 0.11649 -0.29461 65.55 -0.33911
0.792 2 0.11864 -0.29461 71.07 -0.33634
0.88 2 0.12049 -0.29326 77.97 -0.33247
0.968 2 0.12233 -0.29293 83.49 -0.32915
1.056 2 0.12387 -0.29124 89.70 -0.32473
1.144 2 0.12572 -0.29057 95.91 -0.32030
1.232 2 0.12756 -0.28955 101.43 -0.31643

Table 6.21. Example Data from the Uniaxial Strain Test

Time LVDT #2 LVDT #4 Axial Confining Radial


(sec) Axial Axial Load Pressure Deform_ion
Deform_ion Deform_ion (N) (kPa) (mm)
(mm) (mm)

0 0.13894 -0.23056 266.934 8.28 -0.22792


0.078 0.13679 -0.23191 88.978 8.28 -0.22792
0.156 0.13709 -0.23124 151.263 8.28 -0.22792
0.234 0.13863 -0.23056 275.832 8.28 -0.22792
0.312 0.13956 -0.23023 338.117 8.28 -0.22792
0.39 0.14109 -0.22854 444.892 8.28 -0.22792
0.468 0.14263 -0.22719 542.768 8.28 -0.22792
0.546 0.14417 -0.22585 667.336 8.28 -0.22847
0.624 0.14540 -0.22450 774.110 8.28 -0.22847
0.702 0.14601 -0.22382 800.803 8.28 -0.22847
0.78 0.14724 -0.22248 907.576 8.28 -0.22847
0.858 0.14847 -0.22113 1023.247 8.28 -0.22847
0.936 0.15001 -0.21944 1165.611 8.97 -0.22847
1.014 0.15093 -0.21809 1263.486 8.28 -0.22847
1.092 0.15155 -0.21708 1290.180 8.97 -0.22847
1.17 0.15278 -0.2154 1414.753 8.28 -0.22902
1.248 0.15401 -0.21405 1530.424 8.97 -0.22902
1.326 0.15493 -0.21236 1637.197 8.28 -0.22902
1.404 0.15555 -0.21135 1663.890 8.97 -0.22902

156
6. 5.4 Material Properties and Material Property Model

Figure 6.3 shows that the test data from the the frequency sweep, simple shear,
volumetric and uniaxial strain tests are input to the material property model for each asphalt
content. The frequency sweep is used to determine the linear viscoelastic properties (i.e.
complex modulus) and the parameters of the power law. When the log of the complex
modulus is plotted against the log of the frequency, the slope of the resulting line gives the
parameter S. This S can be related to another mixture property, m, which is the slope of the
log creep compliance curve. The parameter m is used in both the calculation of the fatigue
cracking (in determining a Paris' law coefficient) and the permanent deformation.

Data from the volumetric, uniaxial strain, and simple shear at constant height tests
are used concurrently to determine the elastic or resilient and plastic (Vermeer) properties
of the mixture. The resilient (elastic) properties of the asphalt concrete are determined from
the components kl, k2, k3, k6; the Poisson's ratio is determined from the components k4 and
k5. The plastic components (_, Z, _bp,d_cv)are used in the Vermeer model in the
determination of the permanent deformation characteristics of the asphalt concrete.

Table 6.22 illustrates the computed material properties developed from the test data
input to the material property model. These properties relate to the resilient and plastic
parameters identified previously.

Referring to figure 6.3, the input through the Superpave software to the
Environmental Effects Model is the following:

Average minimum/maximum pavement temperature for each day of the year.


Rainfall, solar radiation, cloud cover and wind speed based upon geographic
regions.

6. 5. 5 Environment Effects Model

The output from the Environmental Effects Model is divided into blocks of similar
temperatures defined as a season. The minimum season length is defined as one month.
The temperatures are calculated as average pavement temperatures (C) at one-third depth
for each season. Table 6.23 illustrates example average pavement temperatures at one-
third depth for estimated seasons calculated with the Environmental Effects Model.

157
i

m _ t--
.p=q
Z..
_ , _ , _ ,

_ L

0 _ g

o_

m _ _ _ m. t',xl. _

em

158
Table 6.23. Average Pavement Temperatures at Depth Calculated by the
Environmental Effects Model

Number of Months in Season Average Pavement Temperature (C)


at 1/3 Depth
2 5.91
1 13.22
1 20.98
2 29.55
3 31.63
1 22.54
1 15.53
1 9.38

Total = 8 seasons

6.5.6 Response and Distress Models to Predict Rutting

Superpave allows the mix designer to identify and input the thickness of the total
pavement structure. The designer also provides the input for the material properties for all
layers (subgrade, subbase, base, etc) below the asphalt layer. These material properties are
entered as "default values" which are moduli values. The material properties (table 6.22)
calculated by the Material Property Model are interpolated to the average pavement
temperature at 1/3 depth for each season. The finite element model associated with the
response model then calculates the stresses and strains in the asphalt layer associated with
the material properties of the structural layers and the rate of traffic loading.

The calculated stresses and strains are then used by the distress model to predict the
amount of rutting or permanent deformation. Table 6.24 shows example results provided as
output from these two models. The traffic identified in table 6.24 is the traffic estimated by
the response and distress models at the end of the seasons calculated by the Environment
Effects Model. The rut depth is expressed in millimeters. This same series of calculations
would also be made for the two other asphalt contents as discussed in chapter 5, and would
be extended to the prediction of fatigue cracking and (or) low-temperature cracking as
necessary for a given project.

159
Table 6.24. Example of Predicted Rut Depth Data from Distress Model

Estim_ed Traffic (ESALs x 10 6) Rut Depth (mm)


0.08333 0.65
0.12500 0.86
0.16667 0.50
0.25000 0.74
0.37500 0.93
0.41667 1.02
0.45834 1.02
0.50000 1.02

6.5. 7 Selection of the Optimum Asphalt Content

The procedures discussed previously are followed for the remaining asphalt contents
identified in chapter 5 and the rut depths are predicted. The predicted values of rut depth
(permanent deformation) are used to select the optimum asphalt content. The magnitudes
of the predicted rut depths are plotted against the asphalt contents. Figure 6.5 illustrates the
graphical plot. Table 5.3 of chapter 5 presents a suggested acceptable level for rutting of
10 mm. This value is shown in figure 6.5 on vertical axis as the acceptable level and is
projected to the right to intersect with the plot. At this point of intersection a vertical line
is projected perpendicular to the horizontal axis as shown in figure 6.5.

Asphalt contents to the left of the vertical line in figure 6.5 as shown by the shaded
area satisfy the rutting or permanent deformation requirement. The optimum asphalt
content may be selected within this range identified by the shaded area. If a range of
acceptable asphalt binder contents cannot be established from which to select an optimum
asphalt content that satisfies the rutting or permanent deformation requirement, the mix
proportioning must be adjusted or a modifier included in the mix design.

160
Figure 6.5. Selection of an Optimum Asphalt Content Range Satisfying Permanent
Deformation (Rutting) Requirements

16.0

12.0

_- 14.0 ACCEPTABLE _
:=:
rt I0.0 _,. , . , .- .. r _ " "I." ." ;.-

,,, _'.. .:.': '." .."... o,/


8.0 : _ , 4 w / 6

z _' : ,', , %/ :. :
_ , ' a /'.r ., .
: _f_ "0 I @ . *

o.o ,., , .,/..


.@ 4 ._ " i_
. _ :,, "

_.',.._'_,.. "..
,o * P I
"."

" I .I O I I 0 _ . If. I
20. " ," # . . ' ", " '' :/
I'"

_" ""
r
"
_
I

"
B. 8
") t I ,,
5.5 6.0 6.5
LOW DESIGN HIGH
ASPHALT CONTENT, C%

161
Appendix A Superpave Aggregate Gradation Control
Points and Restricted Zones
Table A-1.37.5 mm Nominal Maximum Size

Control Point (Percent Passing)

Sieve Size Minimum Maximum

75 _tm 0 6

2.36 mm 15 41

25.0 mm - 90

Nominal maximum (37.5 mm) 90 100

Maximum (50.0 mm) 100 T

Table A-2. 25.0 mm Nominal Maximum Size

Control Point (Percent Passing)

Sieve Size Minimum Maximum

75 _tm 1 7

2.36 mm 19 45

19.0 mm 90

Nominal maximum (25.0 mm) 90 100

Maximum (37.5 mm) 100

Table A-3. 19.0 mm Nominal Maximum Size

Control Point (Percent Passing)

Sieve Size Minimum Maximum

75 _tm 2 8

2.36 mm 23 49

12.5 mm 90

Nominal maximum (19.0 mm) 90 100

Maximum (25.0 mm) 100

163
Table A-4. 12.5 mm Nominal Maximum Size

Control Point (Percent Passing)


Sieve Size Minimum Maximum

75 _tm 2 10
2.36 mm 28 58

9.5 mm - 90

Nominal maximum (12.5 mm) 90 I00

Maximum (19.0 mm) 100

Table A-5. 9.5 mm Nominal Maximum Size

Control Point (Percent Passing)


Sieve Size Minimum Maximum

75 _tm 2 10
2.36 mm 32 67

4.75 mm 90

Nominal maximum (9.5 mm) 90 100

Maximum (12.5 mm) 100

164
r--: r-: r--
N ai , _ _ _ ._.

<
E

- _ o_ " _

-_- _ _ - _

" _ _ , _ _ _ -_.

_ _._

'4," .,

[" ._ m _ ,-

165
Appendix B Superpave Specimen Numbers, Test Codes
and Temperature Codes for Level 2 and
Level 3 Mix Designs
Tables B-1 and B-2 present the quantity of test specimens and their code numbers required
for each performance test in the level 2 and level 3 mix design procedures (exclusive of the
specimens required for the initial volumetric design procedure). The specimen code
numbers, and the test and temperature codes, correspond to those employed by the
Superpave software in guiding the performance testing and analysis. Each table presents the
requirements for one asphalt binder content; generally, a level 2 or level 3 mix design will
evaluate specimens prepared at a minimum of three asphalt binder contents.

Table B-1. Requirements for Superpave Level 2 Mix Design

Specimen Quantity and Code Numbers at


Test Temperature (C) / Temperature Code

Test Test -20/A -10/B 0/C Tefr(FC)/H Te_PD)/G


Code

H Volumetric - -

J Uniaxial Strain - -

S Simple Shear at Constant Height - - 1, 2 1, 2

F Frequency Sweep at Constant - - 1, 2 1, 2


Height

T Indirect Tensile Strength (50 - - 3, 4


mm/min Loading Rate)

C Indirect Tensile Creep 5, 6, 7 5, 6, 7 5, 6, 7

X Indirect Tensile Strength (13 - 5, 6, 7


mm/min Loading Rate)

R Repeated Shear At Constant Stress - 8, 9


Ratio

167
Table B-2. Requirements for Superpave Level 3 Mix Design

Specimen Quantity and Code Numbers at


Test Temperature (C) / Temperature Code

Test Test -20/A - 10/13 0/C 4/D 20/E 40/F Te_PD)/G


Code

H Volumetric - I0, I1 10, 11 I0, 11 -

J Uniaxial Strain - I0, 11 10, 11 I0, 11 -

S Simple Shear at - 12, 13 12, 13 12, 13 -


Constant Height

F Frequency Sweep at - - 12, 13 12, 13 12, 13 -


Constant Height

T Indirect Tensile Strength 14, 15 - 16, 17 18, 19 -


(50 ram/rain loading
rate)

C Indirect Tensile Creep 20, 21, 23, 24 26, 27,


22 25 28

X Indirect Tensile Strength 20, 21, 23, 24, 26, 27,


(13 mm/min loading 22 25 28
Rate)

R Repeated Shear at - 8, 9
Constant Stress Ratio

168
Appendix C Standard Method of Test for Determining
the Percentage of Crushed Fragments in
Gravel 1

C.1. Purpose and Scope

This method of test covers the procedure for determining the percentage of crushed
fragments contained in a sample of crushed gravel.

The percentage of crushed fragments specified in Publication 408 is the minimum


amount considered necessary to provide adequate strength and stability in subbase and the
required voids ratio in bituminous and portland cement concretes.

C.2. Sampling

Obtain the sample in accordance with PTM 607. It is recommended that the washed
and dried aggregate resulting from the procedure of PTM 100, "Amount of Material Finer
than No. 100 or No. 200 Sieve in Aggregate," be used. Select a test portion of the sample
in accordance with PTM 625 and of sufficient size to be not less than the weights specified
in the following table. Do not select samples of an exact, predetermined weight.

Gradation Approx. Min. Weight


lb
#8 0.5 1.0
#57, #67, #7 1.2 2.5
2A, #3, #5, 0GS 2.5 5.0

C.3. Apparatus

Sieves must have square openings of the sizes required in section C.4 conforming to
be requirements of PTM 117, for sieving the samples in accordance with section C.4.

A general-purpose balance is also needed, which meets the requirements of


AASHTO M 231 for the weight range in which the principal sample weight (see section
C.2) is located.

_Pennsylvania Test Method No. 621, reproduced with permission of the Pennsylvania
Department of Transportation.

169
C.4. Procedure

Separate the material on the #4 sieve. Discard all material passing the #4 sieve.
When determining the crush count of AASHTO #3 material, separate on the 1/zinch sieve
and discard all material passing.

Determine to the nearest gram the weight of the crushed fragments in the sample
obtained above. A crushed fragment is one having one or more fractured faces. Nicked
fragments will not be considered as crushed. A fractured face is a face that exposes the
interior of the gravel particle.

C.5. Calculations

The percentage of crushed fragments shall be determined as follows:

A xlO0 =%Crushed
B

where: A = weight of crushed fragments

B = total weight of sample retained on the sieve in section A.4.

C.6. Report

Report the amount crushed to the nearest percent.

C.7. References

Pennsylvania Test Method No. 117


AASHTO M 231

170
Appendix D Superpave Paving Mix Designation
Within the Superpave mix design method, paving mixes are designated based upon the key
characteristics of: nominal maximum size; location of the gradation above or below the
restricted zone; traffic level; and intended location in the pavement structure.

An example paving mix designation is

100-C-12.5-B

where

100 = traffic level expressed in millions of 80 kN ESALs;

C = below the restricted zone, coarse;

12.5 = nominal maximum size, mm; and

B = paving mix located below 100 mm depth, base course.

The possible mix designators are presented in tables D-1 through D-4.

Table D-1. Traffic Level Designator

Designator Traffic Level (ESALs)

0.3 < 3 x 10 5

1 3 x 105 to 1 x 10 6

3 1 10 6 to 3 x 10 6

10 3 X 10 6 to 1 x 10 7

30 1 x 10 7 to 3 x 10 7

100 3 10 7 to 1 10s

300 > 1 x 108

Table D-2. Gradation Designator

Designator Gradation

C Coarse, below the restricted zone

F Fine, above the restricted zone

171
Table D-3. Nominal Maximum Size Designator

Designator Nominal Maximum Size

9.5 9.5 mm

12.5 12.5 mm

19.0 19.0 mm

25.0 25.0 mm

37.5 37.5 mm

Table D-4. Paving Mix Depth Designator

Designator Depth of Paving Mix


blank Within top 100 mm of pavement

B Base, below top 100 mm of pavement

172